The Art of Revision

A student once asked me how many drafts I'd written of Lost in Lexicon, and I had to admit I wasn't sure. Ten, maybe? Fourteen?

Steven King starts a new book with only a situation in mind - people trapped in a car by a rabid dog, for example. If we're to believe him, he doesn't plot. He's not sure where he's going when he starts; he just lets the situation play out on his word processor. Every day, he writes at least two thousand words, straight through the end. When he's done, and has set the manuscript aside for a while until he's ready to revise, he comes in with his weed whacker. For him, the most important goal of revision is cutting his word count by ten percent. Meeting this goal assures him that he's tightened the manuscript and moved the story along. Then he's done. Two drafts and he's done.

For me revision is more complicated. For the past few weeks I've been revising Lost in Lexicon's sequel, which is called, at least for now, The Ice Castle. I'm on about the fourth revision, not counting a constant looping back to choose a more felicitous word or shift a sentence for better flow. This is fewer revisions than my last two books, which I think means I'm getting a sense of what a sensible sequence of revisions looks like, at least for me.

I do plot. I don't make a detailed outline, but I jot down notes as I think of them. I have a sense of where the book is going when I start, and I stay aware of structure as I write. I'm not one of those headlong writers who can keep scribbling forward no matter what. I fix some things as I go along.The words have to sound right to me. I have a strangely hard time writing until I'm sure my main characters' names fit them well.

Once the first draft is done I have something readable and grammatical that basically gets me through the story I wanted to tell. I go back to polish and fix it up before I share it, but that's not a full rewrite.

With each of the three children's books I've written so far (the third, a non-Lexicon book called The Beechwood Flute is next in line for finishing), I've taken this polished first draft through an online writing group called Critique Circle. Strangers and critiquing partners read the work chapter by chapter and give feedback. Are sentences too long? Dialogue too stilted? Is the action hard to visualize? Does the chapter drag? Is there too much description? Not enough? Do the characters' emotional responses convince? In response, I write my second draft.

Meanwhile, spending a lot of time with the manuscript and readers' responses, I begin to diagnose structural problems. I find holes in the plot, characters with weak motivation, sections that take too long or seem somehow skimpy. I weigh the manuscript in my hand and divide it into sections. I get a sense of whether the sections - beginning, middle, and end, or perhaps time in one character's voice versus time in another's - seem to have the right heft in relation to one another. Then I cut a chapter down to a few paragraphs, or drop a scene her and add one there. By now I'm probably calling this a third draft.

Now comes the time to make the language more precise. I work to add sensory detail and distinctive gestures to show emotion. I tighten dialogue and strive to give each character a consistent and distinctive voice. I try to root out cliches, although there are always some left over to be purged in the next draft. I try to start and end each chapter with some image or thought or bit of speech that will stick in the mind. Each chapter needs to be a whole, something complete in itself. I struggle with questions of chapter "hooks" and "cliffhangers." Do my readers need to be cajoled to keep reading, or will gimmicks put them off? I want to create authentic suspense and forward motion that arises from readers who care about the characters These are the matters I wrestle with in a fourth draft.

Once this draft is done, off it will go to my editor, who will find more cliches, vague description, awkward wording, and sketchy characterization. We'll fight about it for a while, and then I'll concede that in most places he's probably right. I'll plunge into draft five. But by then I think I'll be really close. This time, I don't think it will take fourteen drafts.