How to finish the first draft of a novel when the middle sags

Most first drafts do not die at the beginning. Openings are exciting, and a writer can live for months on the energy of a good first chapter. They do not die at the end either, because a writer who can see the end usually runs for it. They die somewhere around the forty percent mark, in the long flat country where the setup is finished, the payoff is far away, and every scene feels like a chore you assigned yourself in a better mood.

I keep a notebook of abandoned openings, and if I am honest it is really a notebook of abandoned middles. The openings were fine. What I never worked out, the first several times, was how to keep going once the book stopped pulling me along. This piece is about that stretch: why it happens, and the practical habits that get a draft from the sagging middle to the last page.

Why middles stall

A middle sags for a few predictable reasons, and naming yours helps more than general encouragement.

You have run out of planned material

Many writers start with a vivid opening situation and a rough sense of the ending. Between them is a gap labeled "things happen." When you reach the gap, you have to invent the book in real time, which is slower and scarier than writing what you already imagined.

The stakes have gone quiet

In the opening, the question of the novel is fresh. By chapter eight or nine, you may have answered small questions without raising bigger ones. If nothing new is at risk, scenes start to feel like furniture.

You know too much and too little

You now know your characters well enough to see what is wrong with the early chapters, but not well enough to know what they will do next. The temptation is to go back and fix the beginning. That feels productive and is usually a way of hiding.

The book changed and you did not notice

Somewhere in the first half the novel turned into a different, often better, book. Your outline still describes the old one. Every scene you write against the old plan feels false, because it is.

Write the ending early

The single change that helped me most was writing a version of the last chapter before the middle was done. Not a polished one. Two or three thousand words, rough, written as if the book already existed.

An ending gives the middle a direction. Once you know where the people end up, you can ask of every middle scene: does this move them toward that place, or away from it in a way that will matter? It also changes your mood. A draft with a written ending feels nearly finished, even when it is not, and that feeling is worth a great many motivational quotes.

If you are unsure what kind of ending your book wants, it helps to look at the shapes other novels have used. Beatrix Hallam's survey of seven kinds of last page is a good place to start, and Clementine Oduya's piece on reading the last page first makes the case that knowing the end does not spoil the pleasure of getting there. That is as true for writers as it is for readers. You will almost certainly rewrite the ending later. The point of writing it now is to have a lighthouse, not a final draft.

Bracket and move on

The second habit is simple and a little embarrassing: when you hit a scene you cannot write yet, you write a note in square brackets and keep going.

[SCENE WHERE THEY ARGUE ABOUT SELLING THE HOUSE. SHE WINS, BUT IT COSTS HER.]

[SOMETHING HERE ABOUT THE BROTHER'S JOB. LOOK UP WHAT A RAIL YARD FOREMAN ACTUALLY DOES.]

[THIS CHAPTER IS TOO SHORT. NEEDS A SECOND BEAT. COME BACK.]

Brackets do three things. They keep momentum, because the scene you are stuck on stops blocking the twenty scenes after it. They keep research from eating your writing time, because you note the question instead of disappearing into it. And they turn a vague dread ("the middle is bad") into a list of specific jobs you can do later, one at a time.

Use a consistent marker, such as a square bracket at the start, so you can search for all of them when the draft is done. Some writers color them or put them in capitals. Whatever you choose, trust it. A draft full of brackets is still a draft. A perfect half of a novel is not.

Targets: words or time

People argue about word counts versus time, and the honest answer is that both work for different temperaments.

A word target (say, 500 or 750 words a session) suits writers who like visible progress and can tolerate writing badly. It is easy to measure and hard to argue with. Its weakness is that it rewards padding and punishes days of hard thinking.

A time target (an hour, or ninety minutes, phone in another room) suits writers who think slowly and revise as they go. It counts staring at a stuck scene as work, which it often is. Its weakness is that an hour can pass in rereading.

What matters more than the number is that it is small enough to hit on a bad day. A target you miss three times a week becomes a source of guilt, and guilt is a poor fuel for a novel. I would rather see a writer do 300 words every day than 2,000 words on Saturdays and nothing in between, because the daily writer stays inside the book. The people and their problems stay warm in your head between sessions, and that is half the battle in a middle.

A reverse outline at the halfway point

When you reach roughly the middle of the draft, stop for one session and make a reverse outline. Not an outline of what you plan to write, but of what you have actually written.

Go chapter by chapter, and for each one write a single line: what happens, and what changes. If nothing changes, write that down too. You will learn several things quickly:

• Which threads you have dropped without meaning to.
• Which characters have disappeared for six chapters.
• Where the same kind of scene keeps repeating (three dinner arguments in a row, say).
• What the book seems to be about now, which may differ from what you set out to write.

Then write one paragraph, no more, describing the second half as you now see it. That paragraph is your map for the rest of the draft. It is not binding. It is just better than the outline you wrote six months ago for a different book.

Permission to write badly

Every craft book says this, and it is still the hardest advice to take. A first draft is not meant to be good. It is meant to exist, so that you have something to revise.

The trouble is that writers who read well know exactly how bad their sentences are. The gap between what you can recognize and what you can produce is widest in a first draft, and the middle is where it hurts most, because you no longer have the excitement of the opening to carry you over it.

A trick that helps: write the middle as if it were notes for a novel rather than the novel itself. Allow summary where you would normally dramatize. Allow "and then they talked for an hour about the money, and it went badly." You can expand later. Very often the summary version tells you which scenes deserve full treatment and which never did.

It is worth remembering how slowly real books are made. Salvatore Scibona's first novel, The End, took him about ten years, and he has said in interviews that it began with almost nothing: the sound of a man's footsteps on a staircase and a door at the top. Our profile of the novelist traces how that small image grew into a whole Ohio neighborhood. A draft does not need to know where it is going on day one, and a slow middle is not proof that the book is failing.

A draft that stalled at chapter nine

Here is an invented case, built from patterns I have seen in my own drafts and heard about from other writers.

A writer, call her Marta, is drafting a novel about two adult siblings who return to their late mother's house in a river town to clear it out before a sale. The first eight chapters went well: the arrival, the first night, the discovery of a box of letters that suggests their mother had a second family. Marta knows the ending. The siblings will part on bad terms, and the house will burn, though not by either of their hands.

At chapter nine she stopped. For six weeks she opened the file, reread chapter eight, changed some commas and closed it.

Here is how she got moving again.

1. She named the problem. In a reverse outline of chapters one to eight, she noticed that every chapter turned on a discovery in the house. The letters, a photograph, a locked drawer. Chapter nine was meant to be another discovery, and she was bored of them. So, presumably, would the reader be.
2. She changed the engine. Instead of the house revealing things, she let a person do it. A neighbor who knew the mother arrives in chapter nine with her own version of events. The middle stopped being a treasure hunt and became an argument.
3. She wrote the ending. Three rough pages: the fire, seen from across the river, and the sister driving away without calling her brother. Writing it made Marta realize the brother had to be the one who trusts the neighbor, so the siblings split over her. That decision shaped every remaining middle chapter.
4. She bracketed the hard parts. The scene where the brother confronts the neighbor became [CONFRONTATION. HE ASKS THE QUESTION SHE HAS BEEN AVOIDING. SHE ANSWERS A DIFFERENT ONE.] She wrote around it and came back two weeks later.
5. She set a small target. Forty-five minutes a day, before work, counted by a kitchen timer. Some days produced two pages and some produced a paragraph. In ten weeks she reached the ending she had already written.

The draft that came out of this was messy. Chapters eleven and fourteen were nearly the same scene. A minor character changed names halfway through. Nine brackets remained. But it was a whole novel, and a whole novel can be revised in a way that half of one cannot.

A short checklist for the stuck middle

When the draft stops moving, run through these in order:

• Make a reverse outline of what exists. Look for repetition and dropped threads.
• Ask what is newly at risk in the next chapter. If the answer is nothing, raise the stakes or cut ahead.
• Write a rough version of the ending, if you have not already.
• Bracket any scene you cannot write today and skip to one you can.
• Shrink your daily target until you can hit it on a bad day.
• Allow summary. Expand later.
• Do not go back and revise the opening until the draft is finished.

The last point is the one I break most often and regret every time. The opening will change anyway once you know the whole book. Fix it then, with the full draft in front of you, and keep the middle moving now. More practical pieces on getting a manuscript done live on our drafting shelf.