Socratic questions for writing: a ladder, not a list

October 20, 2026

Search for Socratic questions for writing and you will find long lists. Fifty questions, a hundred, sorted into neat categories with names like "clarifying" and "probing." Most of them are good questions.

They still fail in a real classroom. Here is why.

A question can do two things to a young writer. It can open a door, or it can make them stumble. The same question does both, depending on when you ask it. "Why does this moment matter to your story?" is a wonderful question for a student who already knows what happens in the moment. For a student who does not, it is a trap. They do not have the answer, so they guess at what you want to hear, and now they are writing for you instead of for their story.

So the problem is not the list. The problem is the order. What I use now is a ladder.

What makes a question Socratic

First, a quick definition, because the word gets stretched.

A Socratic question is one where the thinking stays with the student. You are not quizzing them on something you already know. You are not leading them toward your own idea for their story. You are asking about something only they can answer, and then you wait.

That last part matters more than the question itself. The wait is where the thinking happens. If you fill the silence with a hint, you have answered your own question, and the student has learned that waiting you out works.

A good test: after the student answers, could you have predicted the answer? If yes, it was probably a quiz question in disguise.

The ladder

Here are the four rungs, from the bottom up. The questions on each rung are the exact ones I use with my own class this year. I wrote them down in plain words on purpose, because many of my students write in a second language and a clever question that trips them up is worth less than a plain one that moves them forward.

Rung 1: What is happening?

This is the floor. Before a student can describe a scene or explain why it matters, they need to know what happens in it. I teach my students that each section of their story is one scene, and a scene has a simple shape: someone wants something, something gets in the way, and by the end something has changed, even a little.

So the first questions are about that shape:

• "What does your character want here?"
• "What gets in the way?"
• "What's different by the end?"

These sound almost too simple. They are not. A student who can answer all three has a scene. A student who cannot answer the first one is not ready for anything higher, and asking higher will only make them guess.

Rung 2: What does it look like?

Once a student knows what happens, the next problem is that they tell it instead of letting the reader be there. "It was dark and scary and he was afraid." Every teacher has read that sentence.

The questions on this rung pull the student into the senses, one detail at a time:

• "Where is your character standing right now? What's the first thing they'd notice?"
• "What's one sound in this place?"
• "What's one thing here that tells us how your character feels, without saying the feeling?"

Notice what these do not ask. They do not ask the student to describe everything. A student told to "add more description" usually adds adjectives. A student asked for one sound usually finds the one sound that matters.

Rung 3: What changed?

Now the student has a scene and can see it. This rung asks them to look at the scene from the outside: what is different about the character, or the situation, at the end compared with the start? This is where cause and effect lives, and it is the rung where a story starts to feel like it is going somewhere.

A student can only answer this well if rungs one and two are solid. If they cannot say what the character wanted, they cannot say what changed when they got it or lost it.

Rung 4: What comes next, and why does it matter?

The top rung looks forward. The end of a section points to the next one, and a strong "what comes next" tells your reader who it is about, what problem they face now, and what question the next section has to answer.

• "Now that this scene is over, what does your character have to do next?"
• "What could go wrong?"
• "What does your reader want to find out next?"

This is also where "why does it matter" finally belongs. Asked here, it is not a trap. The student has built the scene, seen it, and watched it change. They have something to stand on.

Why the order matters

If you take one thing from this article, take this: climb from the bottom.

When a student stalls on a question, the instinct is to ask a better question at the same height, or a deeper one. Usually the right move is the opposite. Step down a rung. A student who cannot tell you why a moment matters can almost always tell you what their character wanted in it, and once they say that out loud, the higher question gets easier.

This is the same idea I wrote about in helping a stuck writer find the first sentence: the blank is in the thinking, not on the page. The ladder is just a map of where the blanks usually are.

It also explains why one student in a conference races ahead while another, given the same question, gives you the same answer twice. They are on different rungs. If you want the theory behind that, it is Vygotsky's zone of proximal development: a question only works when it lands in the space a student can reach with help.

When a question is not enough

Here is something the long lists never mention. Sometimes a student answers your question with "I don't know," and they mean it literally. Not "I don't feel like it." They do not know what a scene is, so asking what their character wants in this one is like asking someone to fix a part they have never seen.

Asking again does not help here. When a student does not have the concept, I teach it plainly, in one or two sentences. Then I point to a place where they already did it, in something they wrote earlier, so the example is theirs and not mine. Then I go back to the question.

I avoid handing them a model paragraph. A finished example is very easy to copy, and a copied paragraph teaches nothing. Their own earlier sentence is a much better example, because it proves they can already do the thing.

The last rung is the student asking

The point of the ladder is not to keep climbing it with you forever. The point is that one day the student asks the questions themselves.

You can tell it is working when a student asks the question before you do: "What does she want here?" That is the goal of the whole ladder.

So the support should come down. Early on I explain, point, and ask. Later a short reminder is enough. Later still I only ask. Eventually I say nothing at all about that rung, because the student does it on their own. A question you are still asking in May, the same way you asked it in September, has become a crutch. (I wrote more about this test in what scaffolding in writing really means: a scaffold that never comes down is not a scaffold.)

Where Scaffold fits

I built Scaffold because I could not sit next to every student at once. Its AI, Socrates, works this ladder with each student while they write. In my class it teaches in my words, the same explanations and questions above, and when it points to an example, it quotes the student's own earlier writing. It never writes the story. The student writes every word, and the questions stay theirs to answer.

Try it tomorrow

You do not need any software for this. Next time you conference with a writer:

1. Ask the bottom-rung question first: "What does your character want here?"
2. If they stall, step down a rung or teach the idea in one plain sentence, then ask again.
3. Point to something they already wrote that does it well. Their sentence, not yours.
4. Write the four rungs on a card and leave it on their desk. Over time, notice which rungs they stop needing you for.

Start at the bottom rung. Most stuck writers are stuck lower than we think.