LLM Book Knowledgebase

The Craft of Revision

9 min read

The Craft of Revision

Writing and revision are often treated as sequential phases — first you write, then you revise. The more accurate picture, advanced by both William Zinsser and John McPhee, is that revision is writing: the act of first drafting is merely externalizing a thought so that it can be seen clearly enough to be improved. The draft is not the product; it is the medium through which the product becomes visible.

Zinsser: Rewriting Is Where the Game Is Won

Zinsser states the foundational claim directly:

“Rewriting is the essence of writing well: it’s where the game is won or lost.”

And more expansively:

“You won’t write well until you understand that writing is an evolving process, not a finished product.”

This reframing is important for writers who experience drafts as assessments — as evidence of their ability or lack thereof. The draft is not an assessment; it is raw material. The writer’s job in the first draft is simply to get something into existence that wasn’t there before. The writer’s job in revision is to find what is actually worth saying within that material and to say it with maximum clarity and force.

Zinsser’s practical revision process has several layers:

Structural revision: Is the piece organized so that the reader can follow the logic from beginning to end? Is the lead doing its job? Does each paragraph do one thing and do it completely? Is there a clear ending that gives the reader a sense of resolution?

Sentence-level revision: Is every word doing new work? Are there qualifiers and adverbs that can be cut? Are the verbs active? Are there passive constructions that obscure who is doing what?

Sound revision: Does the piece sound right when read aloud? Are there rhythms that drag? Sounds that clash? Sentences that all move at the same pace and therefore put the reader to sleep?

The test for every revision decision:

“Look for the clutter in your writing and prune it ruthlessly… Is every word doing new work? Can any thought be expressed with more economy? Is anything pompous or pretentious or faddish? Are you hanging on to something useless just because you think it’s beautiful?”

The last question is the hardest one. Writers become attached to the most ornate passages — the ones that felt best during drafting, the ones that demonstrate the most effort. Zinsser’s counsel is relentless: if it doesn’t serve the communication of the idea, it goes, regardless of how much work it represents.

McPhee: The Draft as Nucleus

McPhee quotes an explanation he has returned to repeatedly when teaching — a letter he wrote to a student named Jenny:

“Dear Jenny: The way to do a piece of writing is three or four times over, never once. For me, the hardest part comes first, getting something — anything — out in front of me. Sometimes in a nervous frenzy I just fling words as if I were flinging mud at a wall. Blurt out, heave out, babble out something — anything — as a first draft. With that, you have achieved a sort of nucleus. Then, as you work it over and alter it, you begin to shape sentences that score higher with the ear and eye. Edit it again — top to bottom. The chances are that about now you’ll be seeing something that you are sort of eager for others to see.”

The “nucleus” metaphor is precise. The first draft has mass — it exists, it can be revised — but it does not yet have form. The revision process is the process of giving it form. Without the nucleus, the thinking that produces the form has nothing to attach itself to.

McPhee’s most important insight about revision is that most of it happens away from the desk:

“What I have left out is the interstitial time. You finish that first awful blurting, and then you put the thing aside. You get in your car and drive home. On the way, your mind is still knitting at the words. You think of a better way to say something, a good phrase to correct a certain problem. Without the drafted version — if it did not exist — you obviously would not be thinking of things that would improve it. In short, you may be actually writing only two or three hours a day, but your mind, in one way or another, is working on it twenty-four hours a day — yes, while you sleep — but only if some sort of draft or earlier version already exists. Until it exists, writing has not really begun.”

This observation has structural significance: the mental processing that produces the best revisions is not voluntary. It happens automatically once a draft exists, in the background, during sleep, during transit, during activities that free the conscious mind from deliberate effort. The implication is that the single most important thing a writer can do is produce a draft — even a terrible one — as early as possible in the process, because only then does the subconscious begin its revision work.

“You can’t make a fix unless you know what is broken.”

This McPhee maxim is the epistemological foundation of revision. The first draft reveals what is broken. The writer who delays drafting in order to think more carefully about what they want to say is delaying the diagnosis of the actual problems, which can only be identified from the outside — from reading one’s own prose as a reader rather than as a thinker.

McPhee: The Revision of Structure

McPhee also addresses the specific challenge of structural revision — revising not the sentences but the architecture of the piece:

“Often, after you have reviewed your notes many times and thought through your material, it is difficult to frame much of a structure until you write a lead. You wade around in your notes, getting nowhere… So stop everything. Stop looking at the notes. Hunt through your mind for a good beginning. Then write it… Writing a successful lead, in other words, can illuminate the structure problem for you and cause you to see the piece whole.”

This is the productive use of the lead as a diagnostic tool. The lead is not just the beginning of the piece — it is the piece’s theory of itself, its claim about what it is trying to do. Once the lead is written and working, the structure of the rest often becomes visible.

The revision of endings is equally specific:

“If you have come to your planned ending and it doesn’t seem to be working, run your eye up the page and the page before that. You may see that your best ending is somewhere in there, that you were finished before you thought you were.”

Writers consistently overshoot their endings. The actual moment of resolution often occurs earlier than expected, and the material that follows it is a kind of nervous explanation that dilutes the effect. Cutting to the actual ending, which has already occurred, is one of the most reliably effective revision moves.

Godin: Revision as Iteration, Not Perfection

Seth Godin’s model of revision in The Practice shifts the frame from craft to process. He is less interested in the mechanics of sentence-level revision and more interested in the cycle of shipping and learning:

“Every creator who has engaged in the practice has a long, nearly infinite string of failures. All the ways not to start a novel, not to invent the light bulb, not to transform a relationship. Again and again, creative leaders fail. It is the foundation of our work. We fail and then we edit and then we do it again.”

For Godin, revision is the iterative loop of the creative practice: ship, observe, revise, ship again. The purpose of revision is not to achieve perfection before release — perfectionism, he argues, is a form of resistance:

“Getting rid of your typos, your glitches, and your obvious errors is the cost of being in the game. But the last three layers of polish might be perfectionism, not service to your audience.”

The distinction is important: correcting what is clearly broken is an obligation; polishing past the point of good enough is avoidance. The work that is “not working (yet)” is best served by shipping it, observing what doesn’t work for the audience, and revising based on real feedback rather than theoretical anticipation of failure.

“Instead of saying, ‘I’m stuck, I can’t come up with anything good,’ it’s far more effective to say, ‘I’ve finished this, and now I need to make it better.‘”

This reframing — from “I have nothing” to “I have something that needs improvement” — is structurally the same as McPhee’s nucleus. The draft exists. Revision has something to work with.

The Revision Trap

There is a real risk of using revision as a form of avoidance. The writer who revises indefinitely is not improving the work — they are postponing the moment of exposure. The healthy revision cycle has an exit: a point at which the writer determines that the work is as good as they can make it within the available time, and ships it. Both Zinsser and Godin acknowledge this, though they approach it differently: Zinsser from the craft side (know when the piece is done), Godin from the practice side (shipping is the point, not the last resort).

The Subconscious as Co-Reviser

Both Zinsser and McPhee note that some revision is involuntary:

Zinsser: “Your subconscious mind does more writing than you think. Often you’ll spend a whole day trying to fight your way out of some verbal thicket in which you seem to be tangled beyond salvation. Frequently a solution will occur to you the next morning when you plunge back in. While you slept, your writer’s mind didn’t.”

This observation has a practical implication: the writer who works until total exhaustion and then returns to find the problem still unsolved may simply need to stop working and let the subconscious continue. The subconscious reviser is not on call — it works when the deliberate mind is disengaged. The routine of regular work and regular rest creates the conditions under which it operates most productively.

Backlinks