Let the Audience Understand a TV Protagonist Without Excusing Their Harm
Let the Audience Understand a TV Protagonist Without Excusing Their Harm
The client says the flow is exactly right and says Ines's name while saying it. There is a beat where she could correct the attribution. She says thank you, and then she explains the reasoning behind the flow — the part about how it handles the client's old data without a rebuild — and she explains it fluently, because she heard it explained at a desk three days ago by the person who worked it out. Tobias isn't in the room. Ines decided the client would be easier with a small meeting, and she told him so.
That night, her own account of the afternoon is coherent. The client has shelved two ideas from junior chairs already this year. The client trusts her. If the flow had arrived with Tobias's name attached, it would have gone on the pile, the studio would be re-pitching the account in the spring, and the work — his work, which is good — would have died. She means to correct the record once the direction is approved.
This is a scene from a show that doesn't exist. I've invented it for this piece, and the studio, the client, the episode, the people in it are all fictional — nothing here is drawn from a real production, and no scene like it has been shot or tested with anyone. The problem it poses is real enough, though, because it turns up in pitch documents constantly: a lead we're supposed to stay close to does something plainly harmful in act one, and the pitch has to describe the show's relationship to that act without either flattening her into a villain or quietly asking the reader to sign off on what she did.
The usual escape is a phrase. Morally gray. Complicated. We understand her even when we don't agree with her. None of that is checkable, and none of it tells a reader anything about what the show will do.
What is checkable is access. Who does the audience travel with, what does that person know, and what happens to the harm after the episode stops being about the lead's feelings about it. Understanding a person is cheap to produce — a rationale, a history, a pause before the thank-you — and agreement is a separate question entirely. The craft decision is not whether to make her legible. It's what the story holds alongside her account, and whether it holds it long enough for the harm to remain a claim rather than a mood.
What She Says It Was For, and What the Scene Shows
Write the rationalization first, and write it as though it were true, because it may be substantially true. Ines is right that this client is dismissive of junior staff. She's right that the account is fragile. A rationalization that collapses the moment it's spoken isn't a rationalization; it's an excuse with the seams showing, and the audience disposes of it in one scene.
Then separate it from the events. Three facts are on the record here regardless of what she tells herself: she chose not to bring him, she let the attribution stand, and she performed his reasoning as her own. Whatever the reason, those happened. The reasoning doesn't get to rewrite the meeting; it only gets to explain why someone with a working conscience would do it anyway.
The next move is to give the harm a name that isn't a feeling. Tobias hasn't just been slighted. At a client-facing studio, the client's knowledge of you is the thing that determines what you're trusted with — it's standing, not sentiment. When the client assigns the next phase, they'll assign it to the person whose flow they think they saw. That's the loss, it's concrete, and it's the sort of loss a pitch reader can feel the weight of without being told to.
Two traps sit on either side of this. The first is treating a sympathetic history as a settlement. If Ines built this studio from almost nothing and has been talked over by clients for a decade, that belongs on screen — and it still doesn't decide what the room scene means. The second is treating an unpleasant personality as the answer. She can be short with the intern and impatient in reviews; that's character, not verdict. Neither biography nor irritability settles the interpretation, and a pitch that leans on either one is usually trying to skip the scene.
There's also a distinction worth keeping clear in the document itself: an intended audience relationship is a plan, and an outcome for every viewer is not. A showrunner saying we wanted the audience to hold both things at once is describing an intention. It becomes evidence of anything only when someone asks actual viewers, and then it's evidence about those viewers, on that day, in that cut — not a property of the script. I don't have a showrunner's account of audience alignment to check this piece against, and no version of this episode has been made, so the reasoning here is craft reasoning about a constructed example. That's a real limit, and it means the useful work is scene-level: not the audience will understand she's right about the client but in the meeting, Ines does not correct the attribution, and the client leaves believing the flow was hers.
Two Ways to Cover the Same Meeting
Take the same afternoon and allocate attention two ways.
Treatment A stays inside Ines through the meeting. We get the pause before the thank-you, the half-second where she starts a sentence about where the idea came from and finishes it somewhere else, the particular relief of praise landing on a lie you've told well. We watch her elaborate on his reasoning. We never enter Tobias's experience of the meeting, and we don't cut to his face.
Treatment B keeps the act — we watch her do it — but withholds her interiority in the room, and then commits the episode's remaining attention to Tobias: what he does with the fact, and what he asks for next.
The obvious, lazy third option isn't a structure of its own; it's A with one exception, and it's worth naming so you can avoid it: keep the meeting inside her head and cut to Tobias's face for a second while it happens. That shot feels generous. It isn't. His face becomes a caption for her conscience — a visual note that reads she knows — and it organizes him entirely around her interior state. The evidence he provides in that frame is about her.
Now compare what each treatment actually carries.
A gives you the machinery of self-justification in real time. That's not a small thing; it's the scene's whole reason to exist if the show's premise is that this person is worth following. It also gives you a nasty, precise pleasure that distance can't reproduce: the audience watching someone build her account of an event while the event is still happening. What A withholds is any account of the harm from the person it landed on, and any check on her reading of the client. We have only her word that the client shelves junior work. Even if the studio's history backs her up, the scene hasn't tested the claim.
B gives you a claim with standing. Tobias, given his own scene, can want something; a want is a thing the story has to answer, in a way that a feeling isn't. B also lets the harm be understood by someone other than its author. What B withholds is the function of the meeting itself — if we never learn why she didn't say his name, we're watching a person do a bad thing for undisclosed reasons, and the pitch's premise, that she's someone we travel with, quietly erodes. And distance alone doesn't buy clarity. If Tobias never meets the client, B still hasn't tested whether Ines was right about them; it has only moved the camera to a different kind of ignorance.
Neither treatment wins on shape. What A buys is the mechanism of rationalization, and what B buys is a person who can contest the account. Both are worth having, and the choice is where you spend them — often in different scenes rather than the same one.
Which is worth saying plainly: a close version can carry its consequence without ever widening. Stay with Ines, and let the client's follow-up email arrive asking one technical question about migrating the old data — a question only Tobias can answer, which means her next scene is standing at his desk asking him for more, while he has just read a message thanking her for her flow. She gets what she wants. She also acquires a debt she can't name, and every subsequent scene tests whether she pays it. Understanding of her accumulates through choices rather than arriving as a verdict. That is a legitimate series, and it's not an evasion. Its cost is that the audience's picture of the harm stays a proposition — real, but recited only in her voice.
Give the Harmed Person Something to Decide
Three questions do most of the work on the character on the receiving end. What do they want? What do they know? What can they do about it?
Wanting is where most drafts go wrong, because they supply feeling and stop. Distress is a state; it doesn't point anywhere. I want the client to know who designed the flow before I build the next phase is a direction with an object and a date attached, and it can be said out loud in one sentence without becoming a speech about integrity.
Knowledge is where you can get some free asymmetry. Tobias knows because the client's email copied the studio. He doesn't know why she did it. That gap is useful: it lets him act on the fact of the misattribution without the audience or the episode having yet supplied her reason, which means his action stays about him rather than about her psychology.
Capability is the part pitches skip. Ask what changes if the lead refuses. If the answer is nothing much, the harmed character has been given screen time rather than agency, and the audience will eventually notice that his objections are decorative. In this example, the next presentation is nine days out and its core is his work. If he withholds it, the studio doesn't have a deck. That's leverage that shows up in logistics rather than melodrama, and logistics is usually more convincing.
The sharpest way to give another character the ability to challenge the lead's version isn't to give them an argument. It's to give them a fact. Tobias doesn't have to say you're a liar or deliver a mini-address on credit. He can say the email says her flow. The document sits there and can't be talked around — and the lead now has to either correct the record or build on top of it. A lecture lets everyone off the hook because it's about values. A fact forces the next scene.
Checking What Survives the Apology
Here is the failure version, and note that it survives in both treatments, which is why proximity isn't the real question.
In B, we stay with Tobias for twelve quiet minutes: the email, the walk home, whatever he does with it when there's nothing to do. Then Ines finds out he knows, understands what she did, and apologizes — and it's a good apology, and he's moved by it, and the studio agrees he'll be credited in the next deck. The working arrangement is restored. What happened to his decision? There wasn't one; the apology got there first. His distress was fuel for her scene. That version is worse than a plain failure of proximity, because the distance has been spent to make her apology land harder — the wider camera has been recruited into her arc.
In A, the same failure takes a shorter route. She feels the weight of it over the next two scenes, she corrects the record, and the correction is played as her growth. The correction is entirely hers: the client learns who designed the flow because Ines decides to say so, on her timing, in her framing, and Tobias's standing is handed back to him by an act he had no part in. The loss is repaired. He still never decides anything. Tobias appears once, thanking her, and makes no demand.
So the test for either is two questions. At the end of the episode, what is different that the characters can't undo? Not what punishment landed — the demand for punishment is a different rule and usually a worse one, since a lead can get away with real damage and the show can be honest about it. And second: whose choice produced the difference? A's failure version is exactly why the second question is needed, because by the first one it looks handled — the record is corrected, the client knows, the standing loss is made good. What you're looking for is whether the consequence the scene made important is still there, and whether it is still in the hands of the person it landed on. Did the arrangement between them change because he moved it? Does someone else know now, and did either of them choose that? Does the debt recur in a later episode in a form that isn't a feeling? Unresolved is fine. Unexamined is the problem — and so is a problem the lead has quietly resolved on someone else's behalf.
This is also where the pitch document has to be corrected rather than the character. If the logline says the series examines both sides, then the episode needs a scene where the other side decides something, or the logline is a claim the script will embarrass. If the pitch says the show stays with Ines throughout, don't spend a paragraph promising a balance the episodes don't deliver; the reader will assume the writer hadn't noticed. And if the description currently says morally gray lead, delete it. It transmits nothing. Replace it with the scene: In episode three, Ines takes credit for a colleague's design, and the client never learns. That, a reader can argue with — and that's the point. A description that can be argued with is a decision. A phrase that can't is a vibe.
The Version I'd Draft First
I'd keep the meeting close. Ines's hesitation is the scene's evidence, and if the pitch is selling a person worth watching, that's what you're selling: not her likability, her legibility. The audience should get to watch her decide, in real time, to accept something that isn't hers, and to watch her hear herself explain his reasoning without flinching.
Then I'd widen, in a separate scene, and I would not let the widening depend on her regret. Tobias has already read the email when she comes to his desk. He answers the migration question, because the project needs the answer and he isn't trying to punish anyone. Then he says the other thing, and he says it the way you'd state a date: he'll build the next presentation's core, but not unless the client knows who designed the flow. Not a demand for an apology. A condition on his work, which is the only thing he has that the studio can't substitute.
Nine days to the meeting. The deck is his.
Ines gets the last shot of the scene and no line to close it with, which is exactly the position a pitch like this should leave her in: understood, uncorrected, and no longer in charge of the story's account of what happened. Her explanation of the afternoon is still available to her, and it's still not the version the audience is working from. The decision about what to do next doesn't belong to her interiority anymore. It belongs to the scene after this one, where she either goes to the client or doesn't — with Tobias's condition standing there, unaddressed, in the way.
Frequently asked questions
What is the checkable craft question when a protagonist causes harm, and what phrases fail it?
The checkable question is access: who the audience travels with, what that person knows, and what happens to the harm after the episode stops being about the lead's feelings about it. Understanding a person is cheap to produce; agreement is separate. Phrases like morally gray, complicated, or we understand her even when we don't agree are not checkable and don't tell a reader what the show will do.
Why write the rationalization first as if it were true, and then separate it from events?
A rationalization that collapses the moment it's spoken is an excuse with seams showing. In the invented scene, Ines is right that the client is dismissive and the account fragile. But three facts remain regardless: she chose not to bring Tobias, she let the attribution stand, and she performed his reasoning as her own. The reasoning explains why someone with a working conscience would do it anyway; it doesn't rewrite the meeting. The harm also needs a name that isn't a feeling — losing the client's knowledge of who designed the flow affects standing and future assignments.
What do the two treatments of the meeting each buy and withhold?
Treatment A stays inside Ines and gives the machinery of self-justification in real time, including the pause before the thank-you. It withholds any account of the harm from the person it landed on and any check on her reading of the client. Treatment B keeps the act but withholds her interiority in the room and commits the episode's remaining attention to Tobias, giving a claim with standing and letting the harm be understood by someone other than its author. It withholds the function of the meeting if we never learn why she didn't say his name. Distance alone doesn't buy clarity if Tobias never meets the client.
How do you give the harmed person something to decide rather than decorative screen time?
Ask what they want, what they know, and what they can do. Wanting needs an object and a date — for example, wanting the client to know who designed the flow before the next phase is built. Knowledge can supply asymmetry: Tobias knows because the client's email copied the studio, but he doesn't know why she did it. Capability means checking what changes if the lead refuses; if the answer is nothing much, the harmed character has screen time rather than agency. The sharpest challenge is a fact, not a lecture: he can say the email says her flow, forcing the next scene.
What test checks whether an apology has resolved the harm on the lead's terms?
At the end of the episode, ask what is different that the characters can't undo, and whose choice produced that difference. Not what punishment landed. A failure version has Tobias's distress fuel Ines's apology, after which the working arrangement is restored and he never decides anything. Another failure has her correct the record as her own growth, handing his standing back through an act he had no part in. The article says unresolved is fine; unexamined is a problem, and so is a problem the lead quietly resolved on someone else's behalf.