I was just brainstorming a story with a friend, and learned something about myself! (I know, I was supposed to be learning something about the story, but heck, self-knowledge arrives in mysterious ways.)
See, we'd worked out the very basics of the situation and the main characters, and then my friend began speculating about the past of the protagonist, what conflict was plaguing him-- fear of abandonment? Loss of faith?
I was surprised to feel something akin to panic. No, no, I cried. Let's not decide on what his secret is, what his conflict is!
Instead, I wanted just to start writing without knowing much. And then we could look at what the characters were doing and saying and what they WEREN'T doing and saying, and from that, we could figure out what caused what. For example, if it developed that he always diverted questions with a joke, we could then think about what he was hiding. That is, I realized that what I like to DO (rather than what I teach :) is to let my subconscious create the character, and then my conscious can see the actions and reactions and "discover" the secrets and conflicts and internal motivations and all that.
Now this isn't the RIGHT way. There is no right way. What is right depends on the writer. But I'm thinking that maybe it's the way that's the most fun for me. I think it replicates the experience of reading, actually, to observe the character and notice the oddities and use those as a clue to the inner life. Of course, I think what's really going on is that my subconscious has created a character with an inner life already-- there's no conscious invention going on, but that sounds sort of crazy. But maybe not. Just the subconscious can create a dream and finish it in less than a minute, perhaps it can in a burst of inspiration create a whole character. (Or, well, maybe the character exists already... nah. Too weird.)
Contemplating these two different approaches to characterization, I realized they connect with two modes of reasoning. My idea of watching the character and using the "what" (she has panic attacks, she has a Boston accent) to find the "why" (her elderly parents spent their retirement funds to send her to Harvard, and she's never gotten over the stress and fear of failure)-- that's like inductive reasoning-- looking at the particulars and from there deriving a more general principle.
But others start with some sort of premise, idea, or conflict-- say, "I want to write a modern Iphigenia story to explore what it's like being a young woman sacrificed for the general good." What sort of woman would face that conflict? What would her family be like? If her father would sacrifice her, how would she feel about him?
That's more like deductive reasoning-- starting with an idea and from that deriving the way the a character would behave.
No great insight here, but I realize that the inductive is really what works for me, what makes writing fun for me. That process of discovery is what I enjoy. And yet, my analytical side is forever driving me to plot everything out, to invent rather than discover the characters, to control events rather than letting them develop. Neither way is right or wrong, but I think the inductive is right for me.
But that doesn't mean I won't get a chance to analyze, to speculate, to invent... it just isn't at the beginning. In fact, I'm going to try to write until I get stuck-- discover the characters through what I write, and then, when I can't figure out any more, or write any more, I can start analyzing.
So... how do you think you create best? Do you start with the outside-- the behavior, etc-- and move in to the psychology or theme? Or do you start with a more general idea and create the character? Or something else?