Every writer has heard listen to your characters and trust the audience. These are ten stories of great writers who already knew the advice — and the mistake that finally made it mean something.
August Wilson once asked a playwright friend how you make characters talk.
“You don’t,” the friend said. “You listen to them.”1
The advice fits in five words. What let Wilson use it was something else. When he started writing plays, he later said, he couldn’t write good dialogue because he didn’t respect how the Black people around him talked. He thought that to make art out of it, he had to change it.
“Once I learned to value and respect my characters, I could really hear them.”2
That gap — between hearing a piece of advice and having the perception that makes it usable — is what this essay is about.
Screenwriting advice is full of sentences that are easy to understand: Trust the audience. Listen to your characters. Don’t force the story. Find the theme. You can understand every word and still have no idea what the sentence is pointing at. Then a draft fails, an actor surprises you, an audience reacts before you expected it to — and an old sentence means something new.
The test for each story below is strict: did the experience change what the writer would notice on the next script? Because different insights correct different mistakes, some of them contradict each other. Writing advice is medicine, not physics. A principle can be true and still be exactly the wrong medicine for the mistake you’re making.
So don’t only ask is this true? Ask what mistake it became true in response to — and whether you’re making that mistake now.
Sometimes the breakthrough isn’t a new technique. The writer has misunderstood where the problem actually lives.
Woody had to end Toy Story selfless. So, Stanton has said, the team “naively thought” he had to start selfish — you’ve got to start from someplace.3
That produced the wrong character. What worked was different: Woody could be kind, generous, funny and considerate, as long as one condition held — he stayed Andy’s top toy. Threaten that condition and the flaw appears. Now the flaw had somewhere specific to live.
A character flaw doesn’t have to contaminate the whole personality. It can be conditional: a perfectly functional way of being that breaks under a specific threat.
This might click if you think giving a character room to grow means making them miserable to be around at the beginning.
In his mid-twenties, after his marriage and several other relationships had broken apart, Paul Schrader landed in hospital with an ulcer. Talking to a nurse, he realized he hadn’t spoken to anyone in weeks. Then the metaphor arrived — in his words, “this person in an iron box, a coffin, floating round the city, but seemingly alone.”4
He didn’t treat the taxi as a lucky solution to Taxi Driver. It became a method:
“Isolate a problem in your life. Then come up with a metaphor for that problem. Then start writing through that metaphor and see what happens.”5
His discovery wasn’t simply write what you know. It was closer to: don’t write the literal facts of your pain. Find the physical form capable of containing it.
This might click if you know exactly what your movie means, but that meaning still has no body.
In a journalism class at Beverly Hills High School, Nora Ephron’s teacher, Mr. Simms, dictated a set of facts: the principal had announced that the faculty would travel to Sacramento on Thursday for a colloquium on new teaching methods, where Margaret Mead and two others would speak. Write the lead.
The class did what they had been taught. They put the most important-sounding facts first: Margaret Mead and others will address the faculty in Sacramento on Thursday, the principal announced today.
Simms tore them up. The lead, he said, was: There will be no school Thursday.
“It is about the point! It is about figuring out what the point is.”6
Every student had every fact. Not one of them had asked what those facts meant to the person reading. Ephron said she fell in love with that puzzle — sifting through the facts until you find the story, and the truth of the story.
This is find the theme before it becomes a slogan. The facts of a story can all be correct, and correctly ranked by importance, and still miss what the story is actually about.
This might click if you have every event of your story in the right order and still can’t say what it’s about.
Steven Pressfield retells how Chayefsky typed his theme in one line and taped it to his typewriter, so nothing off-theme got in.7 Chayefsky’s own account is less tidy: “You think you have a theme and you then start telling the story. Pretty soon the characters take over and the story takes over and you realize your theme isn’t being executed by the story, so you start changing the theme.”8
The discipline works in both directions: hold the idea still long enough to build from it, then notice when the work has discovered something the idea didn’t.
Competence creates a new danger. Once you get good, you acquire ways to make broken material look functional. You can polish the dialogue instead of fixing the scene, add plot instead of understanding the character, make the behavior more sensible instead of asking why the strange version bothered you. And because the pages stay competent, nothing about them says stop.
In the early design of Breaking Bad, Jesse Pinkman was supposed to die in the first season. His death had a mechanical purpose: Jesse dies, Walt is affected, the machinery continues.
Then Aaron Paul played him. (The 2007–08 writers’ strike also cut that season short, which took the planned death off the table for the moment.) Watching the Walt–Jesse relationship become something the plan hadn’t accounted for, Gilligan said, “it became pretty clear early on that that would be a huge, colossal mistake to kill off Jesse.”9
The lesson isn’t merely that actors can improve characters. A character in an outline and a character played by an actor are different sources of information. Once the played version is richer than the assigned function, protecting the original plan can become blindness.
This might click if you’ve already decided what a secondary character is for, so you’ve stopped noticing who they’re becoming.
Charlie Kaufman spent months unable to adapt Susan Orlean’s The Orchid Thief. The failure seemed external to the screenplay: I have a book, I’m supposed to adapt it, I can’t find the movie. Then the category changed.
“I find I write best when I write what I’m thinking about. What I was thinking about was that I was completely unable to write this script.”10
His inability to adapt the book entered the adaptation. The lesson isn’t “when stuck, put yourself in the movie.” It’s that the obstacle you keep treating as external to the work may contain information about what the work actually wants to examine.
This might click if you keep treating the thing you can’t solve as interference instead of asking whether it belongs in the movie.
While searching for the right structure for 21 Grams, Guillermo Arriaga came across an unfinished novel he had started at 24. It began with a dying man and moved through his life out of order. The old manuscript showed him something he had already done by instinct, before he had a theory for it.11
“Tell the story as we tell our own daily life stories. We never tell them in a straight line.”11
Kaufman found the subject inside his resistance. Arriaga found the form inside an abandoned experiment.
This might click if you know the story, but you keep forcing it through the order you think stories are supposed to arrive in.
Matthew Weiner came to Mad Men from The Sopranos, and the note he took from David Chase is the one he now gives his own writers:
“Take the risk of doing the extreme thing, the embarrassing thing, the thing that’s in your subconscious.”12
What makes it more than a slogan is the moment he describes it coming from. He gives a writer a note — why don’t you do X? — and the writer answers, that’s the thing I wanted to do. “That’s what I learned at The Sopranos,” he said. “That’s the note I try to give to everyone who writes here.”12
The writer didn’t lack the idea. They had it, judged it too strange or too exposing, and swapped in the sensible version. That doesn’t make every embarrassing impulse profound. It means the sensible choice isn’t automatically more truthful just because it moves the plot efficiently. Sometimes skill is what edits the revealing thing out.
This might click if you keep fixing the embarrassing thing your character would do into the sensible thing a competent screenwriter thinks they should do.
William Goldman: “Screenplays are structure.”13 Hayao Miyazaki, who starts production without a finished story: “It’s not me who makes the film. The film makes itself and I have no choice but to follow.”14
One helps the writer whose story has no spine. The other helps the writer whose grip on the spine is strangling everything attached to it.
Eventually the hard thing may not be learning a method. It may be recognizing that a method, instinct or artistic value that has served you well is now keeping you from seeing the material in front of you.
The original ending of Get Out was bleaker. Chris escapes the house only to be arrested by real police; the system outside the horror becomes another version of the horror inside it.
Peele had a reason for that ending, and then the reason expired:
“Early on, the movie was meant to call out racism that was underneath the surface. By the time I started shooting the movie, we were in a more woke America. I realized that the point of this movie no longer was to point out that racism exists. It needed to become about giving a hero, and a release, and an escape, and some joy.”15
The test screening agreed. Producer Sean McKittrick remembers the original ending playing “like we punched everybody in the gut.”16 Audiences reached the police lights already knowing what those lights could mean for Chris. The dread had done the arguing. So Rod gets out of the car.
The lesson isn’t “happy endings test better.” It’s that once the audience has taken the point, continuing to demonstrate it can weaken the story instead of strengthening it.
This might click if you’re keeping the harsher ending to make a point the audience already took.
On Chinatown, Towne met almost the opposite problem. His ending had Evelyn kill her father and go to jail, unable to say why; he found Roman Polanski’s version “excessively melodramatic.” A quarter-century later: “with a story of that complexity, the simpler, more brutal ending is almost the only thing possible… I think Roman was right.”17
Towne changed his judgment; Peele changed the movie. Together they kill a seductive fake rule: darker is not automatically braver, and hopeful is not automatically cowardly. Towne was holding back weight the story needed. Peele was adding weight after the audience already felt it. Same surface decision. Opposite disease.
Krzysztof Kieślowski began as a documentarian, committed to filming real life as it was. The closer he got to the moments that mattered most, the less the method worked. He explained why he left:
“I’m frightened of real tears. In fact, I don’t even know whether I’ve got the right to photograph them. At such times I feel like somebody who’s found himself in a realm which is, in fact, out of bounds. That’s the main reason why I escaped from documentaries.”18
He turned to fiction, and to the intimate moral situations of Dekalog and the Three Colours trilogy, because invented people could go where filming real ones couldn’t.
The discovery wasn’t that the truth doesn’t matter. It was that the literal record is not the only way in, and at the most private moments it may be the way that’s closed. Fiction is how you get into the room.
This might click if you’re holding your story to what really happened, and it’s keeping you out of the moments that matter most.
Paolo Sorrentino spent years developing an unmistakable visual language: elaborate staging, irony, distance, conspicuous control. Then he made The Hand of God, drawn from his adolescence and the death of his parents. The material wanted something else:
“The simplicity of the story, the reality of the family, and the honest feelings of that character imposed a very simple style, something that was unusual for me.”19
It was, he said, the first time he made a movie “without thinking about the style of the mise en scène.” That’s an advanced trap, because his style was never an amateur crutch. It was something he had become exceptionally good at — which is exactly why it could become invisible. A mature voice can turn into a preset: the story arrives, and before it says anything, you already know what your version of it looks like.
This might click if you’ve become so good at your style that you’re applying it before asking whether this story wants it.
A real craft insight doesn’t just tell you what to do next. It changes what you notice next time. After Schrader, an abstract problem looks like a search for a physical metaphor. After Gilligan, a performance can reveal that the function you assigned a character is no longer the character. After Ephron, a correct list of facts stops looking like a story. After Weiner, the embarrassing impulse looks like evidence instead of bad plotting. After Peele, you can notice that the audience has already completed the argument you’re still making. After Sorrentino, you can catch your own voice becoming something you impose.
But an epiphany is not an inoculation. Aaron Sorkin knows intention and obstacle; he teaches intention and obstacle. And he says he needs them to be really solid, “or else I will slip into my old habit, back when I was 21 with the electric typewriter, of just writing snappy dialogue that doesn’t add up to anything.”20
Maybe that’s what mature craft knowledge looks like. Not I learned this, so I no longer make the mistake — but I know what this mistake feels like from the inside.
Sometimes the recognition arrives from within. Often something else has to hand you the evidence: an actor makes the character larger than the outline, an audience reacts before the point is delivered, an abandoned manuscript shows you a form you didn’t know you understood, a style that worked for ten years starts suffocating the material.
Reality keeps submitting notes. Your plan is evidence; reality is evidence too. The hard part is getting good enough to tell when the collision is noise — and when it’s showing you something your plan couldn’t see.
Most writers have a sentence like Wilson’s stored away somewhere. You heard it. You understood the vocabulary. You may even have repeated it to someone else. Then one day the relevant mistake becomes visible, and instead of thinking I know that, you think: Oh. That’s what that meant.
An outside read of your script is one more piece of evidence, like an actor or a test audience. Treat it the way these writers treated theirs: not as a verdict, but as something to check your plan against.
Analyze your script