I blogged earlier about those moments when plot twists emerge in the writer's mind. Lately I was prompted to rethink the nature of this process by a fascinating thought experiment in Robert Burton's On Being Certain. Burton uses the analogy of a computer that's been programmed to solve a pharmacological problem by calculating numerous solutions. The program models theoretical drugs, estimates their probable effectiveness, and displays on the computer's monitor only such solutions as are computed to have a high probability of being effective.
The monitor remains dark for months, then lights up with a formula for a new drug that's theoretically likely to work.
Burton's suggestion is that something similar is happening when a novelist abandons a novel, unable to resolve its plot, thinks about other projects for some months, then wakes up one morning with the perfect-feeling resolution rushing into consciousness. The long delay between conscious cause and conscious effect, together with the feeling of mystical certainty that can accompany revelations from our unconscious, prevents our perceiving what's happened as the slow mechanical working out of a problem: the unconscious sorting its way mechanically through endless, fruitless possibilities, and finally pushing the useful ones into awareness.
George Rabasa tells Glimmer Train, “Whenever you don't know something when you're writing, make it up. You'll be surprised how true it is when you check later.”
Burton writes, “When writing a novel, you can feel the difference between writing 'anything that comes to mind' and willful plotting where you consciously reject certain possibilities. When actively thinking, the censoring editor is in the on position; during unconscious thought it is mercifully muted. But this is only a difference of inputted information; some possibilities are consciously rejected while others are encouraged. From a neural network schema, the basic process of hidden layer processing of inputs remains the same – only the inputs have been changed by the conscious editor. Rather than opting for the dubious premise that 'unthought thoughts' represent a different 'way of thinking,' why not consider cognition as a single entity that is subdivided into various ways of being experienced?”
Note that “when you check later,” in Rabasa's phrase, you will likely do so with a strong bias towards finding what you made up is true.