Trace Repetition to Its Source

An editor’s letter, imagined for the purpose, notes the same marginal comment eleven times in a three-hundred-page manuscript: “Remind us why she is doing this.” The author fixes the eleven places, with a line of motive here and a flashback there, and the next reader leaves the same comment in eleven different places. The cause sat upstream. The character’s want was never established with enough force in the first fifty pages, so every later scene had to borrow its motive from a sentence of explanation.

Each instance looks like a separate defect and can be fixed on its own, with some success. A local repair tends to absorb the symptom without touching the source, and the problem often comes back wherever the manuscript next leans on the missing foundation. Structure sets what the reader knows and when. If an early section fails to supply something later sections depend on, every later section either papers over the gap or leaves it open, and the papering tends to look the same each time, which is why recurrence is diagnostic. Suspect first a structure that keeps creating the same demand, and a habit of yours second. Repeated exposition, as one hypothesis, often points to stakes that were never made clear.

Some repetition is designed. In Shakespeare’s Macbeth, the title character, told of his wife’s death, repeats “Tomorrow, and tomorrow, and tomorrow” and the repeated word enacts the drag of days; the test is whether the second appearance does something the first did not. And sometimes a repeated problem really is a habit, a verbal tic or a favorite gesture, to be treated with a search through the manuscript. Repair one thing upstream and see how much of the recurrence disappears. If most of it goes, the cause was structural. If none does, it was habit.

Practice. Take one chapter or section and read it with a single question: where does the same kind of repair appear more than once? Mark each instance, and group them. For each group, write one sentence on what the reader would have to already know or believe for these instances to be unnecessary. Choose the group with the clearest upstream cause, make a single change at that point, and reread to see how many of the local instances have disappeared. Ninety minutes, all told.