If you're paying attention to human interactions - to the gap between who we are and who we think we are, or the gap between what happened and what we remember - you're going to end up thinking (obliquely or otherwise) about what it means to act ethically, and I think that's all to the good.

The biggest experiment there - and I was convinced for a really long time that it was going to fail horribly - had to do with this weird thing I do every now and then. Like everyone else, as a reader there are certain things that really rub me the wrong way in fiction - pet bugbears, let's call them.

For whatever reason, thus far it's been important to me not to write that kind of collection. Which means that I've spent months playing tic-tac-notecard, trying to get the stories in an order whereby stories that are similar in any given way (diction, narrative stance, setting, plot) are separated by others that aren't.

The most basic organizing principle was pretty straightforward, and is frankly pretty common: the shorter of what are by my lights the two most engaging stories goes first, the longer of the two goes at the end, and everything else goes in the middle.

I start every first draft with voice rather than theme or image or even character as such, so it isn't like I'm ever rubbing my hands, cackling, The dad is really going to take it on the chin in this one! Not in terms of any given story, and certainly not in terms of the collection as a whole.

Realism's anxieties are not my anxieties, but I think I've had its tools close at hand all along. It may be that I'm reaching for them more often than I used to. On the other hand, I'm making no promises for the future. The material itself always gets the last word.