on writing

What Do You Do With a First Draft Novel?

It seems sometimes that writing a novel is a never-ending process. And even when you've finished and the first draft that took you months or years to write is sitting at your desk, you still aren't done. Drafts must be revised.

The temptation to do a quick edit and send the novel off is extreme, but it is a mistake. Novels are big. There is a lot of room in them for error, and every first draft will be full of name-changing characters, badly-described settings, and plot holes you could put the Millennium Falcon through. The opposite temptation, to revise, revise, and revise again, is also a bad idea. Sooner or later, on one pass or another, the story will die in your heart, and every subsequent change will be another stake hammered through it. More to the point, you will never send it off and will be wasting valuable time in which you could be writing another novel.

Over the past two novels I've finished and revised, I've evolved a process for getting through an edit. It's slow and tedious, but it fixes most of the mistakes and is usually thorough enough that I don't need a second edit.

First, I take a break. A month, at least, should pass between finishing the novel and editing it. This gives first readers time to get back to me and gives me enough distance from the story that I can read it more objectively.

Then I print it off. Revising from the computer screen is a mistake; I skim, and you do too whether you realize it or not. Clench your teeth and sacrifice a tree on the altar of art.

I make a quick sketch of the book, writing down the theme, summarizing the plot in a single sentence, listing subplots and writing what I'd put on the back cover of the book if I wanted to sell it. This helps me get a clear picture of the story in my head and will also be useful when I go to write a query letter and synopsis.

Then it's time to dig into the manuscript itself. I only want to do this once. Aside from the poor trees, I have a lot of books I want to write, so the first pass had better be the last pass. I want to fix everything, from plot holes to grammar mistakes.

Novels are intimidatingly big, so the first thing I do is block off a scene. How I decide a scene has begun or ended varies, but it generally means I've shifted time, shifted place, or shuffled people so that the main character is interacting with someone new. It doesn't matter. The point here is to bite off a chunk of story I can chew.

I skim the scene and establish its purpose. A scene ought to move the main plot forward, move a subplot forward, tell the reader something significant about the characters, or tell the reader something significant about the setting. Preferably it will do more than one of those. If it does none of those things or only repeats things I've established earlier I draw a big X over the scene. It's toast.

I figure out whether the scene is in the right place. I try not to run subplot or character-building scenes one right after the other and to keep returning to the main plot, but I also try to keep the main plot from happening in big chunks. If the scene is in the wrong place, I move it. If there's nowhere to move it, then I make a list of essential information in the scene and draw a big X over it.

Then I go through the scene paragraph by paragraph (in the case of dialogue, I take it a clump of lines at a time), and, bearing the essential purpose(s) of the scene in mind, I do a quickie evaluation. If the paragraph does not further the plot or subplots, does not establish setting, and does not say anything new about the character, then it gets the big X. If it does one of those things but distracts from the scene's main purpose, I try and find somewhere else to put it. If I can't, it gets the X.

There are now big holes in the scene, sometimes ones that need filling with new and better material. I write new material in the margins. I keep a notebook handy for when I run out of margin.

Finally, I go over whatever's left of the scene and do line edits: tweaking dialogue, fixing grammar mistakes, and checking for consistency. I try to have a list of character names, slang, and made-up words to refer to so I know that I'm using them consistently. If the characters are referring to events in a past scene, this is where I go back to make sure I remembered the past scene correctly. It's also where I delete references to dead scenes and add references to new ones.

I block out the next scene. Rinse, lather, repeat.

The process isn't really as organized as I've laid it out here, but I will do all of these steps on every scene in the novel. Only the order changes, and even then I always evaluate before I start line edits: it saves me editing stuff I'll only throw away.

Only when I'm done will I go back and type all of the changes in. I have a nasty tendency to change my mind, and it's easier to resurrect a crossed-out scene than a deleted one. Since this is my last pass, I want at least one chance to edit new material, which is why I write my new scenes in a notebook and edit as I type them in. Cut-and-paste is not always your friend.

These steps will not work for everyone, but they've helped me face revisions without the overwhelming urge to run and kept me from endlessly revising in search of perfection. Revision is a necessary chore. Face up to it, and you'll end up with a better novel that you didn't have to spend your life on.

Originally appeared in The SF and Fantasy Workshop Newsletter, Volume 25, Number 289.

Contents copyright © K. Feete, 2005. All rights reserved.