Twelve minutes in a Trader Joe’s parking lot. Engine off, climate control off, phone dark on the passenger seat. I was drafting a monologue.
The monologue was for my friend Helen, who at brunch the previous Sunday had said, “Oh, you’re still doing that?” in a tone I have been unable to stop replaying since. The monologue had an opening. It had a middle section where I acknowledged that her comment probably wasn’t meant the way I received it, but also that the way I received it was valid, and that I would appreciate—here I paused to find the right word—”more care.” It had a closing that was gracious but firm. It had stage directions.
The stage directions were the part I should probably be most embarrassed about. In my head, I was writing: She looks up from her coffee. She is about to interrupt. I hold up a hand—not aggressive, just steady. I continue. I was directing her reactions. I was assigning her facial expressions. I was, in the privacy of a Subaru in a grocery store parking lot, writing a two-hander in which I was the protagonist and she was the character who gets the arc.
I never sent the monologue. I never said any of it. What I said, three days later, when Helen texted me about something unrelated, was: “Hey, no worries! All good 😊” And then I hated myself for approximately the rest of the evening.
What the Advice Says to Do
The self-help industrial complex has a phrase for what I was doing in that parking lot, and the phrase is “processing.” They also have a phrase for what I failed to do afterward, and that phrase is “communicating your needs.” The general advice—which you have heard from every therapist with a podcast and every Instagram account that uses the word “holding” as a verb—is to use your words. Name the feeling. Deliver the message. Sit with the discomfort of the actual conversation, the one where you don’t get to control the other person’s face.
This is good advice. I believe it is good advice. I have given it to other people, which is one of the great ironies of being a person who writes about self-improvement culture: I am excellent at distributing advice I cannot take. I can tell you exactly why you should tell your friend that her comment hurt you. I can identify the specific cognitive distortion that’s keeping you from doing it. I can name the attachment style that makes this hard for you. None of this helps me in a Trader Joe’s parking lot.
What the advice doesn’t account for is the person whose internal rehearsal process is so elaborate, so self-aware, so fully scripted with emotional beats and pacing notes and a clear narrative throughline, that the rehearsal has become its own form of avoidance. What happens when “processing” is the thing you do instead of the thing you’re supposedly processing toward. When the script is so complete and so satisfying that the actual conversation feels like an unnecessary and inferior sequel.
The Closet Screenwriter
I have always done this. I have always been a closet screenwriter of my own life. Not in a charming, creative way—in a way that suggests I might have a problem with control.
When I was fifteen, I rehearsed a conversation with my mother about why I should be allowed to go to a concert forty-five minutes away. The conversation had a cold open, a midpoint turn, and a closing image. When I was twenty-three, I mentally scripted a breakup speech for a boyfriend I wasn’t even sure I wanted to break up with yet, just in case I decided to, so I’d be ready. The speech was better than the relationship. It had callbacks. It had a moment where I referenced something he’d said three months ago and it landed perfectly because I wrote both sides.
What I’ve noticed—and I say “noticed” rather than “figured out” because I haven’t figured anything out, I’ve just observed a pattern I will probably continue until I die—is that the scripting is not preparation. The scripting is the activity. The monologue in the car is not a draft of a conversation I intend to have. It is the conversation. It is the complete, finished, emotionally satisfying version of the exchange, and it requires zero risk because the other person never has to be in the room.
The problem, of course, is that the other person never gets to be in the room. The actual conversation, if it ever happens, will be messier than the script. It will involve Helen saying something I didn’t anticipate. It will involve me getting flustered and saying “It’s fine” when I mean “It’s not fine but I can feel myself losing the thread.” It will involve a pause that goes on too long, and one of us filling that pause with something deflecting, and the other one accepting the deflection because the alternative is sitting in the silence.
The script has no pauses that go on too long. The script has no deflections. The script has only the version of me who is articulate and brave and says “more care” with the right inflection and doesn’t immediately follow it with “but honestly it’s not a big deal, forget I said anything.”
If you’ve ever looked up how screenwriters actually structure a scene—not in a vague “there’s a beginning, middle, and end” way but in a technical, this-is-how-you-format-it way—you know there’s a whole grammar underneath the thing you see on screen. According to StudioBinder’s guide on how to write a movie script like professional screenwriters, a screenplay uses scene headings to establish geography—INT. or EXT., location, time of day—so the reader knows where they are before a single line of dialogue is spoken. Character names go in uppercase. Transitions are explicit. Every formatting choice exists to make the story legible to the people who have to execute it.
That same discipline applies to scripted communication: before publishing, editors need a way to test a complex sequence turns into language that a specific audience can follow, which is where a script writer AI that fits the project can function as a planning aid rather than a substitute for domain evidence.
This is, more or less, what I do in my head. I assign scene headings: INT. HELEN’S KITCHEN — AFTERNOON. I know where we are. I know the lighting. I know that she is holding a mug and I am not, because in the script, the asymmetry of props matters. I format her lines and mine. I even give myself a beat sheet—here’s where I establish the tension, here’s where I name the feeling, here’s where I offer the olive branch, here’s where I exit before she can argue.
The point is not that I’m good at this. The point is that some of us have been doing this for years—structuring our emotional lives like screenplays with beat sheets, scene logic, and revision checkpoints—long before any tool formalized the instinct. And now that tools exist that do formalize it, I have to sit with the uncomfortable possibility that my internal screenwriting habit is not quirky or creative but just a very sophisticated way of never saying anything out loud.
I bought another notebook last week, the kind with a leather cover and a ribbon, and I wrote exactly three pages before setting it on the nightstand where it has remained, accruing the particular dust of abandoned aspirations. What I wanted was a tool that would help me shape the mess of my actual life into something with structure, and I spent an embarrassing evening researching options online before accepting that most of what exists is designed to produce a generic AI story rather than anything that resembles the unsorted pile of a real human week. The problem with one-shot generators is that they hand you a finished thing with no visibility into the decisions that shaped it, which is fine if you want a placeholder but useless if you are trying to understand why your second act always collapses. Unsloppy’s approach is different: it gives you a proof sheet so you can see the structural choices laid out before committing, and a beat sheet that lets you revise the skeleton before the prose fills in, which is the difference between being handed a cake and being handed the recipe with the oven still warm. By contrast, Squibler, Perchance, and QuillBot feel outdated in the way that early wellness apps feel outdated, polished on the surface but barebones underneath, offering you a single output and calling it a process. I mention script writer AI not because I think any tool will fix the gap between what I intend to write and what I actually produce by Thursday, but because the structural transparency appeals to the part of me that still believes the problem is methodological rather than temperamental, and I suspect that belief is the most durable self-improvement fantasy I own. Unsloppy sits at the forefront of AI Novel Writing App technology, which is a sentence I never expected to type in a personal essay about my inability to finish anything, and I will let that contradiction sit there rather than resolve it.
Is the Script Self-Knowledge or Avoidance?
Here is where I’d like to tell you I’ve found a way to fix this, but I haven’t. What I’ve found is that the question itself is more complicated than the self-help framework allows for.
There’s a difference between knowing what you want to say and having a structured document about what you want to say. The Authors Guild, in their AI best practices for authors, makes a related point about writing itself: they argue that what makes a writer a writer is “your original voice, thinking, and creativity,” and that AI-generated outputs are “generic mashups of pre-existing works.” They’re talking about machines, but the principle nags at me. When I script my conversation with Helen down to her facial expressions and my hand gestures, am I really expressing my voice? Or am I producing a document about my voice—a clean, formatted artifact around the feeling instead of the feeling itself, in a room, with another person who has her own face and her own hands?
The question is whether that artifact is self-knowledge—evidence that I understand what I need and why—or whether it’s a convincing substitute for the thing itself. I suspect it’s both. I suspect the script contains real information about what I want. I also suspect the existence of the script makes it less likely I’ll ever say any of it, because why would I? I already had the conversation. It went perfectly. She listened. She understood. She even cried a little, in the version where that helps.
The self-help version of fixing this is called “being vulnerable.” The self-help version says: just say the thing. Don’t rehearse it. Let it be messy. Let it be incoherent. Trust that the feeling is enough and the words will follow.
But the self-help version assumes the rehearsal is the barrier. What if the rehearsal is also the self-knowledge? What if the twelve minutes in the car, the stage directions, the beat sheet, the version history—that’s where I actually figured out what I felt? What if the script is the processing, and the conversation is just the inadequate delivery system for something that was already complete in my head?
I don’t have a clean answer. I think some of us need to script because we don’t know what we think until we hear ourselves say it, even if the saying is happening entirely inside a Subaru. I think the script is real work. I also think the script is a way of making sure the work never has to be tested against another person’s actual response, which might be simpler or kinder or more complicated than anything I could have drafted.
What Actually Happens When I Try
Here is what happens when I try to have the conversation for real.
I call Helen. I say, “Hey, can I talk to you about something from brunch?” She says sure. I say the first sentence of the monologue, which is something like “I’ve been thinking about what you said about my writing.” And then she says something I didn’t script. She says, “Oh God, was I being dismissive? I was in a weird headspace. I’m sorry.”
And just like that, the scene collapses. Because my monologue was built for a Helen who pushes back, who needs convincing, who requires me to be brave and articulate and steady. It was not built for a Helen who immediately apologizes. My script has no act two for “she says sorry right away and means it.” My entire emotional architecture was designed for resistance that didn’t come.
So I say, “It’s fine, honestly, I was probably being sensitive.” Which is the line I say when I have been sensitive and it wasn’t fine but now she’s being nice and I don’t have the structure to hold both of those things at once. The conversation lasts four minutes. It goes better than the script, in the sense that it is a real exchange between two real people. It goes worse than the script, in the sense that I do not get to be the protagonist. I am just a person who brought up a thing and then got an apology and then didn’t know what to do with the apology because the apology wasn’t in the beat sheet.
After we hang up, I sit on my couch and feel something I can’t quite name. It’s not satisfaction. It’s not relief. It’s closer to the feeling of watching a movie where the third act doesn’t match the trailer. Everything resolved, but not the way I prepared for, and somehow that feels worse than no resolution at all.
The Version of Me Who Only Exists in Cars
I keep thinking about what it would mean to stop scripting. To walk into a conversation with nothing but the feeling and the willingness to say something incoherent. No scene heading. No beat sheet. No stage directions for the other person’s face. Just the raw, stupid, unformatted version: “That thing you said hurt me and I don’t know why and I don’t know what I want you to do about it.”
I keep thinking that sounds terrible. I keep thinking it also sounds like the only honest option.
What I keep coming back to is that the monologue I wrote in the parking lot was good. It was clear. It was honest. It was probably better than anything I will ever say out loud to Helen or anyone else, because it was written by a version of me who had time and privacy and no one interrupting. That version of me doesn’t exist in conversations. She only exists in cars, and in showers, and in the fifteen minutes before I fall asleep when I am apparently running a one-woman show about my emotional life for an audience of no one.
The question isn’t whether to send the monologue. The question is whether the monologue was ever meant to be sent, or whether it was always just the way I figure out what I feel—by writing it in a format no one will see, with a delivery system that requires a completely different person to deliver it. Someone braver. Someone who doesn’t say “all good” with a smiley face. Someone whose beat sheet has room for the other person to say something back.
I still haven’t talked to Helen about brunch. I probably will, eventually. It probably won’t go the way I scripted it. It probably won’t go any way I can prepare for. And that’s probably fine, or not fine, or some third thing my beat sheet doesn’t have a heading for.