Then there's the draft writer, who leads an admirably dismal existence. He starts the same way every other sensual artist does. He's got characters floating in his unconscious. He intuits their yearning. He has attached to them a milieu, a circumstance, perhaps an external moment in the world, some event to block that yearning. These are the basic elements you all have when you start a novel. The draft writer begins a draft for the very purposes I've been talking about; he is rightly afraid of being drawn into his mind and his analytical self. He would never preplan, because that would trap him like literal memory, like a "message," like preconceived ends, and thereby destroy his ability to get into the unconscious. So the draft writer feels the necessity of taking the merest hints to start the novel and then plunging in, making approximations, writing rough, by any and all means continuing to write and write and write through a great sprawling draft. And the draft writer relishes this. "Ah, I've got this mass of stuff, and OK, I've got to do the second draft now and the third and the fourth, and the seventeenth, and that's fine. " Great works of art have been created this way, and I suspect statistically it's the more common way to write a novel. It's done because those artists understand the danger of being sucked into their heads.
But you know what? They're just deferring the problem. Because once you have this great raw sprawling first draft, how do you find that leaner, more coherent second draft? The dangers of analysis are very powerful in that search.
What I'm suggesting instead is this:
You go to your writing space as you would on a day when you're planning to write words. You go into your trance, just as you would if you were writing your new book sentence to sentence. But that's not what you do. Instead you're going to do what I call dreamstorming—not brainstorming, dream-storming. You're going to sit or recline in your writing space in your trance, and you're going to free-float, free-associate, sit with your character, watch your character move around in the potential world of this novel. You're going to dream around in this novel, one level removed from moment-to-moment writing — that is, at the level of scene. You're going to do this for six or eight or ten or twelve weeks, every day. You're going to go into your writing space, you're going to go into your dreamspace, you're going to float around, and you're going to dreamstorm potential scenes in such a novel as this with such characters as these, with such yearnings as these. And you'll try to float everywhere in the noveclass="underline" beginning, middle, end— all over.
You'll have a pad of paper in front of you (you can do it at your computer if you prefer; I do it by hand on legal pads); you'll make a list. You're going to write down on this legal pad six or eight or ten words, not many more, that represent a potential scene, just identifiers of scenes. Don't hesitate to put something down, as long as it's coming with a sensual hook. You're going to make sure that every scene you list has come to you with some — and it can be very faint, very fragmentary— but some sensual, concrete hook. A little vision of something, a little smell or taste of something, a little sound of something. Do not trust a scene that presents itself to you as an idea. Each scene must have an even fragmentary vision, some sort of sense impression attached to it.
Then you write down the briefest identifier of that scene. For example: Lloyd rapes Anna. Darrell ponders his digging trowel—those were typical identifiers from Countrymen of Bones. On a typical day you'll float among a number of possible scenes from different parts of the book. And when a compelling scene comes to you, you might be visited by the draft writer's instinct — you want to start writing the full scene right away. Don't do it. Resist it. Even if that scene is "Wow! It's vivid. It's got, oh man, it's really almost there." You've got the six- or eight-word identifier and you leave it at that. This is coitus interruptus. You float on.
Now, you might find yourself getting into little runs of scenes. This scene provokes an image of another scene and another, possibly in sequence. Well, OK, follow it; that's great. List each one, six or eight words. But as soon as the run peters out, do not force it, do not try to find what goes next.
This is very important: through the whole six or eight or ten or twelve weeks, you do nothing—and I emphasize nothing—to try to organize, structure, or otherwise manipulate these scenes. You do not even try to reconcile totally contradictory scenes. Lloyd rapes Anna; Lloyd thinks of raping Anna hut doesn't. If you have a fragment of each of those scenes on two different days, don't reconcile them. Put it all down, all that contradictory stuff.
Eventually, the law of diminishing returns sets in, the scenes come more slowly, and one day along about the sixth or eighth or tenth or twelfth week you find yourself with only one scene and you say, "Whoa, I'm finished with doing this."
Now you've got what? A hundred and fifty? Two hundred scenes? You may have three hundred. You're ready to go to the next stage.
Say you have two hundred scenes. You buy yourself two hundred three-by-five cards — not five-by-seven; you only need room for a phrase, and you want them to be easy to handle. Turn the cards horizontal. Write the identifying phrase or set of words in the center of the card. Write one scene per card. Now you have two hundred cards with two hundred scenes.
By the way, a word about three-by-five cards. Functional fixedness can cut both ways. Some of you may have a very strong association between three-by-five cards and an academic thesis or dissertation or other analytical work. If so, you may need to change something about them. If you worked with white three-by-fives in your life of the mind, perhaps you can use a different color card for your creative work.
So you've overcome any possible negative associations and you've got your two hundred scenes on two hundred cards. The next day, you go into your writing space, you clear yourself a tabletop, and you go into your trance. Then you start flipping through your two hundred cards. Every time you look at a card there's a little sense impression that jumps off the card at you: bing, bing, bing. What are you doing? You're looking for the first good scene in the book — the best point of attack. Narratively, this scene will obviously be near the beginning of events but may not be the first chronologically; the story may have already begun. You find this scene, you put it in the upper left-hand corner of that big empty space. Now you flip your cards. You're in your trance. You flip the cards looking for that second scene. What scene would follow the one in the upper left corner of your table? You find it, you put it up there next to the first, and so forth. At the end of the first day, you've got, for example, eight cards in a row. Pick them up in order. Bind them tight.
The next day, you come into your writing space, you go into your trance, you flip those eight cards. You're reading your book. You lay them out again, upper left-hand corner of your table. Now you're looking for the next scene in your cards, and so forth.
Now, there are a couple of ways to go here. Let me deal first with the possibility that you're going to go all the way through to the end of the book, arranging your cards, plotting your whole novel this way. I did that with Countrymen of Bones (and in the process my two hundred cards resolved themselves into ninety-two).
What happens as you move along, in your trance, picking up your cards one after the other? Say you get to card number 22—scene number 22—and when you choose the next scene out of the remaining 178, you realize there's a hiatus.