Schur Farms: The Organic Creation of Some of the Century’s Most Iconic Characters
The writer and showrunner breaks down the science behind Ron Swanson, Leslie Knope, and your favorite ‘Office’ players
There’s been one constant in every round of The Ringer’s Best TV Character of the Century Bracket: characters from Michael Schur’s shows. Schur, a former writer, producer, and (occasional) actor on The Office and the creator or cocreator of Parks and Recreation, Brooklyn Nine-Nine, and The Good Place, had a hand in developing five of the bracket’s contenders: Michael Scott, Ron Swanson, Leslie Knope, Dwight Schrute, and Janet. Round 1 pitted the last two against each other—Janet was marbleized—and Tony Soprano narrowly edged out Leslie in the Sweet 16, but Michael, Ron, and Dwight have made it to the Elite Eight.
Schur, who by popular demand is reluctantly regrowing his Mose beard from The Office for charity, is working on three new TV projects, including Rutherford Falls, an NBC streaming sitcom starring Office alum Ed Helms. With production on his series halted, Schur made time to tell us how to create a great TV character, how classic characters evolve, his TV touchstones, the blessings and curses of deep casts, and the original Ron.
How’s the Mose beard coming in?
Terribly is the answer. Terribly. It’s the worst. I was accused by many people of doing this to get out of shaving my head, and my honest-to-god true response is, I would so much rather shave my head. It’s not even close. It’s a runaway victory for shaving my head over growing this stupid beard.
For you to start selflessly growing an itchy beard at a time when it’s dangerous to touch your face—I don’t want to use the word “hero” lightly, but ...
But go ahead, use it.
Somehow, Mose did not make our bracket of the best TV characters of the past 20 years ...
Then it’s obviously an invalid survey.
… but a bunch of your other characters did! It’s a very scientific process, obviously, both the seeding and the voting, all extremely scientific. Whatever the people say, that is the definitive word on which characters are good.
Yeah, it’s just math. You can’t argue against math.
I read that Rutherford Falls production was postponed. Are you still able to work on anything, besides the beard?
Yeah, so Rutherford was paused. Basically, we were in the middle of the production meeting for the first episode. We had drafts of seven out of the 10 episodes, and two others were broken. So, if it’s possible to be in a good position during something like this, we were in a good position because we were able to, thanks to Zoom, just keep working. We kept the maximum number of people on the payroll that you can, because the whole writers’ room kept working and all the assistants kept working. We’ve just continued to work on the scripts. We meet three times a week and we all have assignments and we go off and work on scripts. Hopefully, whenever this ends we will be just like, “OK, we’ve got all 10 scripts, let’s get going.” There’s other projects that were at different stages. There was a pilot we were going to shoot for HBO Max in late April. That’s obviously completely postponed. There’s an animated show. It’s, like Rutherford, sort of plugging along. We’re making do the best we can. These aren’t real problems. These are fake Hollywood problems.
Since you’re in the early stages of a few shows, this seems like a fitting time to ask about character creation. At what point in the process of creating a show do the characters come to life?
That’s entirely dependent on the show. There are certain shows that are character studies. Your Breaking Bads, let’s say, where you don’t do Breaking Bad if you haven’t understood fundamentally who Walter White is and what’s going to happen. Famously that show was pitched as, “We’re going to take Mr. Chips and turn him into Scarface.” They built a show around the character. It was like, we know who this guy is at the beginning, we know what he’s going to be at the end if we get that far, and everything else fills in around the character.
Then you go the complete other way: Most comedies, I would say, are pretty lo-fi in terms of premise. They’re like a bunch of people hanging out somewhere in an office or in an apartment building in Manhattan. In that case, the discovery of the characters—you have some idea at the beginning. You can’t run a pilot without some idea of like, this is the funny one and this is the snarky one and this is the uptight one. But the characters are built brick by brick slowly by a large group of people over, hopefully, many, many years and hundreds of episodes. You have to know something about the world and something about the characters, it’s just what the ratio is at the beginning of the project.
The Office was being built off of the template from the British show, but there were only four characters who meant anything in the British show. There was David Brent and Gareth, Tim and Dawn, and everybody else was either a two-dimensional cipher or never got developed. When Greg [Daniels] brought the British version to America, he started with Michael Scott, Dwight Schrute, Jim Halpert, and Pam Beesly, and then filled that office with 20 other people. He had some idea of who Oscar was and who Phyllis was, but he very deliberately left them blank at the beginning because it was like, let’s do this organically. Let’s get a bunch of funny people in a room and pitch on, who are these people? What’s their personality trait? How do we learn about them? He knew Angela was a schoolmarm-type uptight person and he knew that Oscar was fastidious, but Oscar didn’t start out as gay. That was a thing that got discovered along the way.
In The Office’s case, it’s like 75 percent setting, 25 percent character to begin with, and then you slowly build the characters over time. With something like Breaking Bad, it’s probably flipped. It’s 75 percent character understanding and 25 percent world-building, and then the world gets built up around the character. It’s much more likely that you have a vast reservoir of knowledge about the character before you begin the project if you’re doing drama, I think, than comedy, but there’s plenty of exceptions to that rule. In fact, The Good Place, I knew way more about those characters before we started than I had on Parks and Rec. Parks and Rec, everybody was vaguely fleshed out in my mind, but then the real work of building the characters was the result of a lot of pitches and a lot of conversations with the writing staff and actors.
You’ve said that most shows typically don’t hit their stride until they’re 10 episodes in, and that it would be nice if you could just toss those first episodes and start over again. How much of that learning curve comes from getting a feel for the characters?
So much. There are like developmental stages of sitcoms. The first developmental stages, the ugly larva stage from episodes one to eight or so where everything is slimy and gross and hard to understand and you don’t even know what kind of animal you’re looking at. Then, at some point, you come through the clouds and you can see things more clearly. The whole first eight, it’s just trial and error. It’s just putting people together in different combinations and pitching jokes and seeing what works and what doesn’t. Then at some point, you hit on something. Greg used to refer to it as the “ur-episode,” where you finally at some point break an episode where you’re like, “Oh OK, this is how people function with each other. This is what’s funny about them. This is how they relate to each other.”
Then the rest of Season 1 is taking that idea and running. Then you hit this real golden period. It’s the sitcom equivalent of when you’re in your 20s and you’re totally invincible and you stay up all night and drink and smoke and then wake up the next morning at 7:00 and go for a run. That’s like seasons 2 and 3 for sitcoms. Seasons 2 and 3, you know who the characters are, you know what makes them fun, how they relate to each other, but also you’re early enough in the proceedings that you can discover new things about them all the time. Everything’s exciting and fun. You make them hook up with each other or you make them do dramatic things and you stumble into really funny stories that are unexpected and you meet their parents. You do all the exciting development off of this strong base that you discovered in those first 8-16 episodes.
That takes you ideally through Season 3. After that, now it’s like the burden’s on you, because you’re comfortably ensconced in adulthood, if you want to continue this developmental metaphor. Now you’re just in adulthood, and to keep things interesting, it just requires a lot of really careful thought and planning and plot creation and character development to keep these characters consistent but also learn new things about them. It just gets harder and harder to do that. But if you really work at it, there’s seasons of sitcoms, Season 5, 6, 7, that are just as good as anything they did earlier.
There’s other wild cards that happen. If you think about Cheers, Shelley Long did that show an enormous service, probably, by bolting, because that show had done however many episodes with her. Sam and Diane is still the standard for a certain kind of relationship story in sitcoms. Then she bolted, and it was like, “Well, how much longer could they have kept going with the will-they-or-won’t-they thing?” They brought in a new woman who presented a completely different sort of challenge for the main character of the show. Sometimes weird stuff like that happens and it helps the shows reinvent themselves.
When you’re in that planning stage, do you come up with comps? Do you start with an archetype or a classic character and say, this is the spin we’re putting on it? He’s like Sam Malone, but bald and in a stable, loving relationship?
You do that sometimes. We’re closing in on 100 years of television. Just hundreds of thousands of episodes. It’s hard to find like a new dynamic, really, between friends or between lovers or between husbands and wives. Once in a while a show does something that’s truly new, I think, with its central relationship. I thought Broad City was like, “This is a new friendship. I’ve never seen this friendship before.” It was so specific, I think because it’s the friendship that those ladies have in real life. But it was just so specific and so modern and so fresh and so interesting that you would just tune in, or at least I would, just to be a fly on the wall of how these two women relate to each other. But usually we don’t have that, so you’re basing it on something.
Ideally you’re not basing it on another television character or characters because everything has been done 50 million times. You’re never gonna find Ted Danson and Shelley Long again. You’re never going to find John Krasinski and Jenna Fischer again. Instead of saying, “This is a Jim and Pam relationship,” you might say, “We can take an aspect of Jim and Pam’s relationship or Sam and Diane’s relationship and weave it in here,” but you just have to get so granular with it and specific with it or else it’s going to seem tired and boring. On Parks and Rec, we talked a lot about Ron Swanson in the early going, not as like, this is Archie Bunker, or this is whoever character from TV, but more like, this is a kind of person who has these specific character traits that we can identify in certain other people both in our lives, and then some on TV.
In that case, I was really sick of people in American politics who called themselves libertarians because I was like, “You don’t really know what a libertarian is.” I was like, what would be fun is to create an actual libertarian, a person who actually lives off the grid and hunts his own food and has an ideologically consistent strain of libertarianism. We just set out with him to be like, let’s make a person who, if he were asked whether gay people should get married, he would say, “Of course they should. The government shouldn’t tell anybody what to do about anything.” I think it’s more frequently stuff like that, where instead of basing your characters on previously seen characters, you’re trying to base them on archetypes or things in the culture or observations that you’ve made about people that you think might be interesting.
How often have you been surprised by the way that certain characters really resonated or failed to connect with an audience? Presumably you start with an internal PECOTA projection for a character. If the early results deviate from that expectation, how do you decide whether it’s a small-sample fluke or a real problem?
I don’t have a feel for what’s going to work and what doesn’t. I really don’t. I think some people do. I think Chuck Lorre seems to have some impressions about what will connect with people or what people will enjoy. That’s a skill I don’t have. I’m much better at saying, I know this actor is really good, so if we can write a good character, this actor will become famous or this character will be beloved. Aziz [Ansari], for example, I was like, that dude’s just deeply funny. He’s just a funny person. I find him really enjoyable to watch. So let’s put him on the show, and if we can design a good character for him and give him funny jokes to say, people will like Tom Haverford.
Then the burden is on the writers to lift the show up to the level of the actor. I felt that way about everybody at Parks and Rec. When Dan Goor and I were starting Brooklyn Nine-Nine, it was like, look, we have Andy Samberg and Andre Braugher. If we screw this up, it is 100 percent on us. If this show doesn’t work, it will not be the fault of Andy Samberg, one of the funniest people in the world, and Andre Braugher, one of America’s greatest living actors. This will be our problem. I don’t know that I ever have accurately predicted one way or the other whether a character will hit. I think with the help of Allison Jones, who’s the greatest casting director in the world, I’m much better at saying, “Oh, I think that actor will work in this show,” and that’s about as far as I get.
If there is a character who changes significantly or gets sidelined—like Sam Seaborn on The West Wing or maybe Mark Brendanawicz on Parks and Rec, or on the flip side, a minor character who ends up stealing the spotlight—how do you handle that?
That’s the human error part of this. The part of it that is organic leads you into weird places that you didn’t expect in a big picture way. You start off with a theory of how things are going to go, and it could be a really good theory and it could make total sense. Then you start executing the plan and it’s like, well, there’s just something not right about this theory. We fed bad information to the computer and it started spitting out a bad result. I remember reading that Family Ties, I think, was basically supposed to be about the parents. Then you have Michael J. Fox come out of nowhere as one of the greatest comedic actors of all time. He and Ted Danson, I think, have the most perfect comedic timing of any actors who have ever existed on television. And so, guess what, Family Ties creators, you shift gears and you say, “This show’s going to be more about Alex P. Keaton now, because we just discovered this guy, and this guy is interesting and he’s brilliant.”
At the beginning of any of these shows, we didn’t know. When Dan and I invented Brooklyn, we thought, well, maybe this character, Amy Santiago, is a potential long-term love interest for Jake, but we didn’t know. It turned out Melissa Fumero is great and she’s super funny and she became exactly what we wanted her to become. But if we had cast a different actor or it just didn’t work, we would have bailed on that. A lot of the stuff you’re talking about, in terms of shifting gears and changing things, the audience never knows about, because it’s just in the heads of the creators and the writers, and then they aim for it and then it just doesn’t work out and they just do something else. What you see is like Plan B or Plan C or Plan D.
To make this bracket, characters had to have great writing and also a great actor portraying them. Is one or the other any less essential?
It’s alchemy at some level. It’s mysterious and complicated and it’s never one thing. Good writing can elevate acting and good acting can elevate writing. Neither one of them is going to work on its own. The best writing in the world in the hands of a clumsy actor will fall flat. The greatest actor in the world without good things to say will fall flat. At The Office once, we were deep in the weeds on some rewrite of some episode in Season 2, and we just didn’t know what to do. Greg Daniels just said, “You know what? Steve will save us.” He basically just punted on whatever the scene was and said, “Maybe Steve will rescue the show,” and of course he did it. He did every week.
In the Season 2 finale of Parks and Rec, the government had been shut down. Dan Goor wrote a cold open for this episode, and at the end of it he wrote, “Pratt does something physical.” That’s all we had. We just left it in, and [Chris] Pratt was on roller blades in that scene. At the end of it, he said something crazy and rolled up to a desk, tried to jump over, and just fell on the ground, and that was the blow of the cold open. There are times when writers just desperately need the actors to save them and vice versa, but the true magic of something really working is always a combination. It’s always a great actor and really good writing and good directing and good costume design. It’s everything.
There’s an infuriatingly ethereal, lightning-in-a-bottle quality to most great television that makes it feel both really special and also impossible to replicate. It’s a lesson I learned at SNL. I think everyone who wants to be a writer should work there for a year, because it is the truest, purest, uncut heroin of the maddening nature of comedy writing. You start on Monday with literally nothing, and on Saturday at 11:30, you do 90 minutes of television and then you start over again. It’s really, really intense and adrenaline fueled. It teaches you a lot of great lessons, and one of the lessons you learn is, you can’t replicate last week, this week. Whatever happened last week, if you had an amazing show and you got a bunch of sketches on the air and they all killed, guess what, you have nothing this week. You’re starting from nothing. You don’t build off what you had. It’s a lot easier in a sitcom because you’re constantly building, and you do have things you can go back to. But the lesson of SNL is really good, which is you have to pay as much attention to what you’re doing every week as you did last week, or else it’s going to fall apart.
How does balancing screen time work on a show like The Good Place, which tended to focus on the core characters, compared to ensemble shows like Parks and Rec or The Office, where you could almost accidentally forget about Phyllis for a few episodes because the cast was so big?
It’s a blessing and a curse. The Office is the craziest example because there were, whatever, 20 series regulars at some point. All the actors knew this, but it was still the case that there were going to be whole weeks where you didn’t say anything or you had four lines and they all got cut. That’s not ideal. There’s some perfect balance of core characters and tertiary characters who you can bring in for comedy. By the end of Parks and Rec, we really had a perfect situation, which is we basically had eight to 10, depending on the year, main characters, all of whom were just dynamos. Then we also had like 50 randos who we could design a situation, a public forum or some plot involving the main characters, and you could bring in whoever, Jason Mantzoukas or Mo Collins, and you just knew what they were going to give you when you brought them in. That’s ideal, I think.
The Office basically took all of those tertiary characters and made them main characters, which is harder to do. There’s going to be whole weeks where Stanley only has a line or two, and that’s not great. The other problem with that is, on Parks and Rec, for example, you’ve got Amy Poehler in the middle of this show, and every second that you’re dealing with some side character is a second that she’s not on camera being Amy Poehler. You want to dance with the one who brung ya a little bit. Sometimes you can get really carried away with funny side characters, and then suddenly you realize, oh, we haven’t given Adam Scott any lines.
I used to think of it as 60, 30, 10. That like, 60 percent of every episode had to be with the real core main characters. Then 30 percent of it could be a B story with the side characters, and then 10 percent of it could be dealer’s choice, Mantzoukas or Mo Collins or whoever. If any less than 60 percent of the main story of a Parks and Rec episode is Leslie, Ron, Chris Traeger, April Ludgate, whoever, you’re in trouble. That means that there’s something out of balance, and you could feel it. You’d have readthroughs where you get to the end, and even if it was really funny and everybody was really great and everything, you’d be like, “Man, we just didn’t write Ron Swanson into a good story this time.” You feel their absence.
Of the characters you worked on who are in our bracket, do any have particularly interesting evolutions or origin stories? Michael became less Brent-like as time went on, for instance.
Yeah, that’s the biggest one. Basically 40-Year-Old Virgin is what did it. Greg made the switch after 40-Year-Old Virgin came out and said, ‘We need to work 20 percent of that guy into this character.’ And we all fought him on it and thought he was stupid and that he was going to ruin the show. Instead, he literally saved the show.
All of those characters had evolutions. The original conception of Ron Swanson, before we ever even wrote the pilot, was that he was corrupt. The very first incarnation, like early, early, early, was that he was a version of a modern-day political animal who is basically siphoning off public funds for his own gain. But I was also so sick to my stomach of that story that you read about all the time. Before we wrote the pilot, I was like, “No, that’s not good,” and that’s when we stumbled on the true, 19th-century-libertarian version. The ur-Episode for Ron Swanson is the episode Dan Goor wrote where Ron has a hernia and basically doesn’t move for the entire episode. He just sits in his chair and refuses to admit that he’s in pain. We shot that whole story in about four hours, because it was just him in his office trying to eat a hamburger by throwing it into his mouth. In the edit room, we were like, “OK, that’s who he is. We got it.”
Janet was going to be a data center, a Star Trek computer-type thing. That was my original conception of her, and then it was like, “Hey, this would be funnier if a human being would portray it,” which is a very obvious conclusion to come to. But we auditioned a million people for that part. The youngest was 12, and the oldest was, I think, 70. That character is the most due to the actor involved. I said to D’Arcy [Carden] early on, this is the deal with who Janet is or the function Janet serves, but I don’t know how you should play this. I know that you shouldn’t speak in any kind of robotic voice, but I don’t really know what you should do. It’s why she was the perfect person to play the part, because she had so much improv training that I was like, “I think you’ve got to go with your gut on a lot of this stuff,” and we started defining Janet based on what she wasn’t, which is why she says all the time, “Not a girl,” “Not a robot,” not a whatever. Do you know that famous story about how Michelangelo carved the David? He chipped away everything that wasn’t the David. That’s basically what we did with Janet. We just kept saying what she wasn’t and then left it in D’Arcy’s hands, and she figured out what she was.
Leslie had an evolution, too. She says in the pilot that she wants to be the president. That was the North Star for us, that she was ambitious and that she was high achieving and had huge dreams and huge goals. But at the beginning, we had her lean into a politician cadence. In the pilot, she walks into a house and Andy’s sitting on the couch and his feet are in casts, and she says, like, “How are you doing, son?” And she goes over to shake his hand. That was us thinking of her as a budding politician instead of as just a true optimist and believer in the power of community and in the power of government to help people. That we didn’t really nail down until like the last episode of the first season. A lot of that credit goes to Amy, because she’s so naturally funny when she just talks and is a human being. She was ahead of us. She kept talking like a real person and we kept trying to write her as a phony politician. Eventually, we were like, “Oh wait, she’s right and we’re wrong.”
Do certain people in the writers’ room have a better feel for certain characters, and if so, will you assign episodes to those specialists if they revolve around those characters?
Definitely, yes. No question. Writers find aptitudes for characters. Alan Yang and Harris Wittels were both Tom Haverford experts. Aisha Muharrar was a Leslie Knope expert. Mindy Kaling was a little bit of a Michael Scott expert. The writers who are experts in characters, it’s not necessarily beneficial to assign them an episode where the character takes center stage. Being an expert in a character means that you can write a joke for them or you just have a fluency in their voice that other people don’t. So they’re going to be around for every rewrite anyway.
Greg had a theory that if you had an episode that was particularly mushy and romantic, don’t give it to a writer who tends to be more excited about mushy and romantic episodes. Give it to the hard-joke writer who has zero tolerance for mushy romance, because it’s just a counterbalance of, make sure you don’t tilt too far in any one direction. A really good episode of a comedy like The Office has a little romance, it has a lot of hard jokes. It’s got a good plot, it’s got some weird little character moments. So if you have an episode where the main story is pointing in one direction, give it to a writer who has a different specialty so that you don’t go off the deep end. I always felt like the real power of the writer who was fluent in a certain character wasn’t in writing the actual draft, it was in the rewrite process. Because then you’d get to a point in whatever draft it was and say, “OK, we need a better joke for Ann here.” And then someone who was really good at writing for Ann would go like, boom, and there it is. Then you get the benefit of everybody’s input.
Are there certain touchstone characters for you in other shows that you keep in mind as examples of what you’re aiming for?
Yeah, sometimes. The characters that I find most interesting are the ones where the actors seem to get it from minute one. Ted Danson and the writing staff of Cheers knew who Sam Malone was in the first frame of the pilot. It was a fully formed character. He knew exactly how to do it. He claims he doesn’t—Ted is a very humble person, and he claims that he was very dissatisfied with his performance or very unsure of himself. I think he’s full of shit, because I’ve watched that show a thousand times. He knows what he’s doing and he knows who his character is from literally the first frame.
There’s definitely episodes of shows that I rewatch all the time. There’s a Friends episode, the one where they played the trivia game. The stakes are who moves into the big apartment. They play girls versus boys trivia. It’s just so wonderful because the show is deep enough in its run where you knew all those characters so perfectly well. Then they found a device to basically just do in-character jokes about each character over and over and over and over and over again, and they’re all funny. There’s really crazy experimental stuff that I go back and watch sometimes. I’ll watch the Breaking Bad episode, “The Fly.” The Lost episode “The Constant,” I watch once a year just because it’s such a masterful piece of character storytelling.
If you ask any writer, they could name 10 to 20 individual episodes of TV that are just real touchstones for them that moved them or affected the way that they thought about their job. I think I watch them not just because I love them or because they help me understand something about storytelling, but it’s just reassuring to watch people execute something perfectly. It’s like watching Mary Lou Retton vault in the ‘84 Olympics or Simone Biles do a floor routine. There’s something incredibly comforting to me about just watching a perfectly executed idea. It’s like, it is possible. It’s possible to be this good, and you should try to be this good all the time.
The history of TV is full of white-guy writers writing, not coincidentally, white-guy characters. One of the main characters on Rutherford is Native American, and roughly half the writers’ room is Native American. In what ways does having writers who share a character’s background help you make a memorable, true-to-life character, and in what ways are there universal traits that go into a great character that are a little less dependent on that lived experience?
I think there are two levels of the benefit. There’s a very straightforward level where you can pitch a story or a joke or a line or an action or anything for a character who isn’t your exact makeup genetically, racially, socioeconomically, religiously, whatever, and if you have a person who does share that makeup with the character, that person can say, “That doesn’t ring true to me. That’s not what this person would do in this situation.” That is a thing that never used to happen, because it was 12 white dudes. It’s happened to me multiple times on Rutherford already. We’re rewriting Episode 4 of 10 now, but in the discussions in the room, I have pitched things for some of the Native characters and some of the Native writers have been like, “Yeah, that’s not what would happen.” It’s like, “Great, I’m glad you’re here. I’m glad to know that, because I have no interest in making something that isn’t authentic or true to life.”
That’s level A. Level B is, they can also pitch things and ideas and actions and lines of dialogue and everything else that not only are true to life, but that you have no access to, and that you would never, in a million years, know to pitch. And those things are interesting and fresh and new and different and fun and exciting. Look, most of The Wire was written by white dudes. It’s not like it’s impossible to put yourself in the mind of a person who isn’t exactly like you. It happens all the time. It should also be noted that in The Wire, a lot of those guys were either novelists or investigative journalists or former journalists or former police officers who knew the world they were talking about inside and out. They at least had spent a lot of time with people in the world they were writing about, which is better than nothing. I wrote lines for Darryl on The Office. Craig is African American, and I’m not, and that doesn’t mean I’m not allowed to do that or that I can’t do that.
But in the Brooklyn writers’ room, in the early going, Prentice Penny had an understanding of Andre Braugher’s mindset that I just straight up don’t. He was very smart and nuanced in the way that he talked about a black man in a position of authority, and Terry [Crews] too, by the way. The insights that he had and that other African American writers had about the African American characters were just things I can’t have. It’s such a reductive and simple and obvious thing to say, but that’s what makes it so infuriating that it was ignored in Hollywood for 80 years. Of course it’s going to be better to have people with multiple viewpoints, and specifically the same viewpoints and touchstones as the characters they’re writing for. How would it not be better to have that?