You sit down to write a paragraph you've been postponing for a week. You stare at the page. Nothing comes. The cursor blinks. You feel a familiar dread that has nothing to do with vocabulary or grammar. What's happening, more often than not, isn't a writing problem. It's that the paragraph would require you to admit something you haven't fully admitted yet. Hayes and Flower's 1980 model of the writing process showed blocks can happen at multiple stages — and Mike Rose (1984) found that the most stubborn ones come from rigid rules the writer has absorbed, not from lack of skill.
You know what you want to say. You have the outline. You have the research. You've written the surrounding sections. The paragraph in question shouldn't be hard. And yet you can't get the first sentence out.
You try a different approach. Different opening. Different framing. None of them work. You make tea. You answer email. You re-read what you wrote yesterday. Hours pass. The paragraph remains blank.
Most writing advice will tell you to push through. Set a timer. Lower your standards. Write a terrible first draft. Some of that helps, sometimes. But there's a kind of block that doesn't yield to any of it, because the obstacle isn't on the page. The obstacle is somewhere underneath.
Some blocks aren't about craft — they're about what the writing would commit you to. Hayes and Flower (1980) and Mike Rose (1984) showed that the most stubborn blocks come from rigid rules and avoidance, not from lack of skill.
What does the research actually say?
The cognitive psychologists John Hayes and Linda Flower published a landmark model of the writing process in 1980. Their model showed that writing isn't a single activity — it's a coordinated set of processes including planning, translating thought into language, and reviewing. Blocks can happen at any of these stages, for different reasons.
A separate body of research, from the educator Mike Rose's 1984 work on writer's block, found something striking. Many serious blocks weren't caused by lack of skill or lack of ideas. They were caused by rules the writer had absorbed — rules about what good writing should be, what shouldn't be said, what would be embarrassing to write. Rose called these "rigid rules" and showed that they often produced exactly the paralysis we associate with writer's block.
The most influential research on this kind of writing comes from the psychologist James Pennebaker. Pennebaker's decades of studies on expressive writing found that the act of putting difficult experience into language — particularly experience the writer had been avoiding — produced measurable cognitive and emotional benefits. The writing itself was the work.
The picture this research draws is uncomfortable. Some blocks are about craft. But many blocks — perhaps the most stubborn ones — are about avoidance. The paragraph won't come because, at some level, you don't want to write it. You don't want to write it because writing it would commit you to a position, surface a contradiction, or admit something you've been circling without naming.
Once the block is recognized as avoidance, the question changes. It's no longer how do I write better. It's what would I be admitting if I wrote this paragraph at all?
Why doesn't "just write through it" work?
Most advice on writer's block treats it as a generative problem. You don't have the words. You need more words. So push, freewrite, lower your bar.
This works fine for the everyday version of the block — the slow start of a session, the mild resistance of a new chapter. It doesn't work for the deeper kind. When the obstacle is what the paragraph would mean, freewriting just produces more avoidance, in higher volume. You can fill pages without ever approaching the thing you're not writing.
Generative AI tools have made this worse in some ways. They can produce text that looks like the paragraph you're stuck on. But that text is rarely yours, and it doesn't resolve the underlying block. The block is about your relationship to the idea. The AI can't have that relationship for you.
What helps with this kind of block isn't another sentence. It's a different question.
What actually helps?
The principle is small: when stuck on a paragraph, ask not how do I write it but what would I have to admit if I wrote it.
Practically, this means having access to a kind of thinking partner who asks questions rather than supplies answers. The questions don't need to be clever. They need to be direct: What's the part of this you're not saying? What would the sentence say if you weren't worried about who reads it? What feels true about this that you can't quite write yet? What would change in your life if you put this on the page?
This is closer to coaching than to writing assistance. Good coaches don't tell you what to think. They ask the question that surfaces what you already, on some level, know.
The British psychoanalyst D. W. Winnicott described something he called the holding environment — a space safe enough that a person can articulate what they've been protecting themselves from. Writing requires something similar. A page that feels safe enough to be honest on. A presence that asks good questions and doesn't judge the answers.
Most AI writing tools don't offer this. They offer to write for you. What's often more useful is something that helps you think.
How Wretify's brainstorming partner works
Wretify's AI brainstorming partner is designed around questions, not output. You tap the microphone and say whatever's on your mind — about the paragraph, about the chapter, about what's been bothering you. The AI doesn't write the paragraph. It distills what you said and asks the one question that might take you deeper.
Each round sharpens the thinking. A writer says, "I can't get past the first paragraph of the identity chapter. I know what I want to say but it won't come out." The AI responds: "The block might not be about the writing. It might be about what you'd have to admit once that paragraph exists. What would it say if you weren't worried about where it leads?"
That's the move. It doesn't give you the paragraph. It gives you a way back into the part of you that already knows what the paragraph is.
I built this because I'd spent years stuck on paragraphs that, in retrospect, were never really stuck. I just hadn't admitted what I was trying to say.
The block isn't in the words. It's in what the words would mean.
- Hayes and Flower's 1980 model showed blocks can happen at multiple stages of the writing process.
- Mike Rose (1984): many blocks come from "rigid rules" the writer has absorbed, not from lack of ideas.
- Pennebaker's expressive writing research: writing into difficult material produces real cognitive and emotional benefits.
- The persistent block is often avoidance — the paragraph won't come because writing it would commit you to something.
- Wretify's AI brainstorming partner asks questions rather than supplying answers, surfacing what you already know.
This article was written in Wretify. Try it free →
Frequently asked questions
No. There are real cognitive and craft causes — being undertrained, having an unclear structure, lacking the research, being physically tired. The point isn't that every block is emotional. It's that the persistent blocks, the ones that don't yield to standard advice, often are.
The boundary between writing and self-understanding has always been blurry. Nonfiction writers in particular are drawing on their own experience and perspective constantly. Acknowledging that doesn't turn writing into therapy. It acknowledges what good writing has always involved.
That's a different problem, and a more solvable one. Research, conversation, reading, and time will usually fix it. The harder block — the one we're describing here — is when you do know what to say and can't seem to write it.
This depends entirely on how the AI is set up. Generic AI tools default to vague, motivational questions. An AI with access to your project and everything you have already written can ask sharper questions because it has context to work with.
That's your choice. Some writers use these sessions as private thinking space and never publish what comes out. Others find that the paragraph that emerges from the block becomes the best paragraph in the chapter. Both happen.