You don't need a dead father to explain a character's sadness. And impressing yourself with wit/cleverness often feels like what it is - authorial intrusion.

To be quite honest, along with thinking and such when it comes to writing, I'm not into words like "theory." I'm a PhD dropout. No matter how many twenty-five-page papers I wrote, I never felt like I was saying much. I didn't feel like the writer of the book, whose work I was analyzing, would have been impressed. It didn't matter how much time or effort I put in.

I've learned a lot about language from reading slush. You can immediately tell if a writer is in control of the narrative. This writer will avoid using too many words like "possibly," "probably," "maybe," "perhaps," etc. He/she will avoid using clichés, as well as a lot of metaphors, and won't take four sentences to say what they could in one (or write a great sentence and follow it up with a bunch of stuff that just weakens it).

I think training your instinct comes from writing and reading. There's no big secret. And reading slush helps, as well; I'd recommend everyone edit a literary magazine at some point. It's time-consuming, but there's a lot to learn from other writers who are also learning. The patterns (twelve stories about whales in this batch?) are also interesting.

There were definitely scenes I struggled with more than others: the car accident and the thunderstorm are two that come to mind. It's difficult to write about a thunderstorm. There are only so many ways to describe it and our vocabulary is so limited. And the car accident scene required a tense, manic quality that had to be conveyed in the language, as well as the character's dialogue and actions. I was editing these scenes long after I thought I was finished with them.

It's difficult for me to imagine any scene differently because I've read the book so many times. The book, as a whole, seems like a document that wouldn't withstand any changes at this point. Or perhaps I simply can't imagine having to revise it again.

I try to think as little as possible, at least while working. I look at some of my early stories and can see the machination behind them, like a gear slowly moving. For example, sticking a dead father into the story to explain a character's sadness and bad decisions, or trying to impress myself with my own cleverness.