Write It Before You Say It: Three Routines for Quiet Secondary Readers

You know the moment. You ask a real question and lean back to let the room think.
A pencil suddenly needs inspecting. A student is suddenly interested in the ceiling tile.
Someone says “I got it,” then a hand goes up, your heart lifts, and the next words are:
“Ms., can I use the bathroom?”

I have been there more times than I can count.

For a long time I read that silence as students not doing the work. Then I paid closer attention to who was quiet, and it was almost always the same students: those currently struggling with grade-level expectations. For them, “share your thinking about the reading” is not a low-stakes invitation. It is a request to figure out what they think, put it into academic-sounding words, and say it in front of thirty people, all at once, in real time. Any one of those is hard. Stacked together with an audience, they become a wall.

I had that part wrong. I kept trying to get those students to talk more. What they needed was for talking to feel less risky.
That is the same move I make with a hard text. I do not lower the text. I strengthen the supports.

The support that changed my room was smaller than it sounds: I have students write the sentence before they say it.

The one move under all three routines

Writing the sentence first breaks the wall down. The thinking and wording happen on paper, privately, where a weak start costs nothing. By the time the student speaks, the hard part is done. They are not composing out loud. They are reading something they already decided was worth saying.

There is a real idea under this.
The Writing Revolution framework is built on a simple premise: writing is how thinking gets built, not just how it gets recorded. Ask a student to write a sentence about the reading and you are not tidying up a thought they already had. You are making the thought exist.

The three routines below are three versions of that one move, and they follow the arc Anita Archer teaches for explicit instruction: model it first, let students practice with support, then gradually remove the support.

Start by modeling it

This is the step I used to skip, and it matters more than anything else.

Put a question on the board and think through your answer out loud. Then show one weak sentence and one strong sentence.

• Weak: “I liked it.”
• Strong: “I think the narrator is lying because on page four he says he was not there.”

Do this before any student tries it alone. Ninety seconds can reset what students think is expected.

You are not fishing for the answer in their heads.
You are showing the move, then handing it over.

Routine 1: The Confidence List, and a script

This is the purest form of the move, and the one I would start with. Everything else is built on it.

Before discussion, students write on their notes:

• One thing you feel confident you could talk about today.
• One quote, detail, or example from the reading they would point to.
• The script line: write, word for word, how they would say their idea or ask their question.

That last line is the whole thing. A student who has written, “I think the narrator is lying because on page four he says he was not even there” is no longer deciding whether to risk speaking in the moment. The risk is already spent. The sentence exists.

When I walk the room while they write, I can see quickly who is ready and who is stuck before discussion starts. The students who usually disappear are the ones I get to quietly coach first.

I saw the first concrete change with Routine 1: students who used to stare at their papers now hand in complete opening lines.

Routine 2: Speaking Cards, deal three and play one

Some students can write a sentence but cannot start one on demand. This routine hands them an opening, and I turn it into a card game because a little play lowers the temperature.

Each student gets a small deck of reusable cards with sentence stems. Here is the important part: they draw three cards and can use the one that fits their thinking.

Card stems I use

• I agree with ___ because ___.
• I want to build on what ___ said by adding ___.
• I am not sure I agree, and here is my evidence.
• A detail that stood out to me is ___.
• This connects to ___ from the reading, and that matters because ___.

The move is the same as Routine 1. A student picks the stem that fits, then writes the full line before speaking. The card gives the first words. The write-first step gives the rest with less pressure.

Routine 2 helped my strongest students too. They had good ideas but needed a faster start, and the stems stopped the “I have nothing useful to say” pause.

Routine 3: Pair first, then public

Before whole-class discussion, students rehearse in the safest room there is: one partner.

• Partners take one prompt and talk through it for sixty seconds.
• Each student writes the single sentence they most want the class to hear.
• In whole-class discussion, students begin with a connect stem if they want one:
  • “Building on what ___ said, I want to add…”
  • “I agree with ___, and I think…”
  • “I am not sure I agree, because…”

If this sounds familiar, it is because it is Think-Pair-Share with one extra step. The added step is the written sentence in the middle.

Without that step, sharing still means composing out loud in front of thirty people. With it, students read a sentence they already own.

One idea I am still testing, and would love for teachers to try: let the partner turn happen in a student’s home language so thinking can happen in the language they own. Then they can bring the idea into English for class.

I have never taught an ENL class, so I am offering this as a direction to test, not a proven formula. If you try it, tell me how it went.

The first two times students try it, it was small and awkward. The third time, I heard sentences come out that looked very close to the ones on the board.

Why this is reading work, not just talk

These routines are not about participation points. They are reading work in another form.
In each routine, students:

• pull a specific detail from text,
• form a claim before speaking,
• build on someone else instead of dropping an isolated comment,
• and practice academic language intentionally.

That is summarizing, citing evidence, and using academic register at the same time.
In a sixty-second partner talk, discussion is not a break from reading work. It is reading work with support turned up where it belongs.

What this fixed for me

Heavy support at the start, less support as students steady.
A student who used to study the ceiling tile starts talking, because the sentence is already in hand.
The next big payoff is confidence, not performance:

• quieter students wrote complete claims before speaking
• more students used text evidence instead of opinion-only comments
• discussion moved from “I am done with this” to “here is my sentence”

Small move. Big shift.
If you try any of these, I’d love to hear what happens. I start rolling this out in September, and your feedback will help me share an even clearer version.