The Right Tool for the Wrong Sentence

You know the sentence is wrong before you know why. It sits on the page and something about it resists — the eye slides off it, or pauses in the wrong place, or the rhythm stumbles where it should flow. You read it again. You read it aloud. The problem doesn't announce itself. It just sits there, wrong. Most writers respond to this by rewriting the sentence. Start over, try a different construction, rearrange the clauses. Sometimes this works. Often it doesn't, because the problem isn't the sentence. The problem is the one before it. I rewrote an opening paragraph eleven times before I realized the issue was the last line of the preceding chapter. It had ended on a note of finality — the character closing a door, the scene wrapped up tight — and the new chapter couldn't start because the previous one had sealed itself shut. I changed the ending. Left the door open, literally and figuratively. The paragraph wrote itself. The tool for this is not talent. The tool is diagnosis. You have to learn to locate the pain accurately. A sentence that feels clumsy in the middle of a paragraph might be perfectly fine in isolation — the problem is what surrounds it. A transition that jars might jar because the scenes on either side are too similar in tone, and the transition is doing work it shouldn't have to do. Here's a diagnostic method I use. When a passage isn't working, I don't stare at the passage. I read the two paragraphs before it and the two paragraphs after it. Five paragraphs total. Almost always, the problem is at the boundary — the exit from one idea and the entrance to another. The passage in the middle is the symptom. The transition is the disease. Another tool: read the problem passage in a different format. Print it. Change the font. Read it on your phone. Your eye has memorized the shape of the words on your screen, and you're reading from memory instead of from the page. A different format forces your eyes to actually see what's there, and what's there is often different from what you thought was there. The hardest diagnostic skill is admitting that the problem sentence might be the one you love most. The one with the image you're proud of, the rhythm you worked on, the word you discovered and slotted in with satisfaction. These sentences are the most likely to be wrong, because you stopped editing them when they pleased you and pleasure is not the same as function. Kill the sentence. You can always bring it back. You almost never will.