In 1928, Maxwell Perkins opened a package from a writer he'd never met. Inside was a manuscript from Thomas Wolfe. Something over 300,000 words. The novel was about everything: childhood, memory, family, the South, hunger, death, the feeling of a particular afternoon in Asheville, North Carolina. Wolfe didn't know how to stop writing. He only knew how to keep going.
Perkins spent months cutting. He removed roughly 90,000 words before Look Homeward, Angel was published. He restructured chapters, carved out entire sections Wolfe loved, found the shape of a novel inside what was really a sprawling accumulation. Wolfe, for a while, was grateful. Then he wasn't. He left Scribner's, told people Perkins had stolen something essential from the book. Wolfe died at thirty-seven, still convinced the best parts of his novel were sitting in a drawer at his old editor's office.
Perkins never said much about it publicly. But the people who knew him understood that he believed the opposite: the novel Wolfe had written was in there all along, buried under the excess, and his job had been to help Wolfe see it.
The tension between those two positions sits at the center of every revision. The writer who feels something was taken. The editor who believes something was found. It's the question you have to answer for yourself every time you reopen a draft.
Structural revision means being willing to break what already works
Michael Ondaatje has described his revision process for The English Patient in a way that sounds, honestly, a little unhinged. He printed pages, cut them apart with scissors, and arranged the pieces on his floor. He'd have sections that worked beautifully on their own, paragraphs he knew were good, and he'd still pull them out of position and try them somewhere else.
What's interesting is that Ondaatje wasn't fixing broken things. He was asking a harder question: whether the pieces were in the right order, whether the reader needed to encounter this image before that conversation, whether a revelation that came early should come late instead. Structural revision requires you to treat your own good writing as movable. The sentence you love in chapter three might belong in chapter nine. It might belong nowhere.
Most writers resist this because it feels destructive. You had a working draft. The scenes connected. Why break something that already functions? But functioning and being right are different things, and Ondaatje seems to understand that a novel can work in a dozen different arrangements, and only one of them is the one that makes the reader feel what you actually wanted them to feel.
The editor sees what the writer can't, and that gap defines the relationship
Robert Gottlieb edited Catch-22. He edited Toni Morrison. His philosophy was deceptively simple: the editor's job is to help the writer write the book the writer wants to write.
That sounds obvious until you think about what it requires. Morrison might hand Gottlieb a draft and say she wants the book to do one thing, and Gottlieb's job was to see where the draft was already doing something else, something she hadn't noticed or hadn't articulated. He had to hold the writer's conscious intention in one hand and the manuscript's unconscious direction in the other, and figure out which one to trust.
Writers don't like being told they don't understand their own book. But the revision process, if you're honest about it, is the part where you keep discovering that you didn't write what you thought you wrote. You wrote something adjacent. Something close. Gottlieb's skill was pointing at the gap between intention and execution without making the writer feel like the gap was a failure.
I'm not sure most writers can do this for themselves. You can learn to see some of the gaps, with enough distance and enough drafts, but there are always blind spots that require another reader.