I think we've all been taught that there's only one kind of revision—you write a draft of a story, and then you revise that story over and over again until it's ready. As it happens, however, there's at least one other possibility. Sometimes that first draft is never going to become a final draft. That doesn't mean it's a waste, though. Sometimes that first draft is actually practice for a different, better story that you're going to be able to write just as soon as you've written enough practice ones.
I've been revising this way from the beginning. My teacher Jesse Lee Kercheval was the first to notice it—she observed that in some cases I would do the traditional kind of revision—first draft, revise, revise, revise, etc., complete—but in other cases I had to write several stories that were about more or less similar things (all about dating, say, or childhood confusion) until I found one that was worth sticking with. This observation was a revelation to me. I'd been worried about how many of my stories were falling by the wayside, but I could see that Jesse Lee was right. Those were practice stories, and I needed to write them not for their own sake but so that I could eventually get the right angle on the material.
This kind of revision is still a big part of my process. Take, for example, my new book The Guy We Didn't Invite to the Orgy and other stories. There's a story in there called "Hunting, Gathering," and it was basically provoked by an experience of being around a person who bothered me. (I get a lot of stories this way.) Ultimately I think the story's main character, who is based very loosely on that person, is complicated, three-dimensional. First, though, I had to write another story called "Cul-de-Sac," in which the main character (based on that same person) is flat and annoying. In other words, I had to write the irritable version of the story before I could write the version that explored its character's full humanity, as any good piece of fiction should.
Or there's "Cowboy," which is about a person who's plopped himself into a foreign cultural context (in this case, the Philippines), where he's going to have to wrestle with his expectations and assumptions about other cultures. Well, before I wrote that story I wrote two others about a similar kind of situation, though those ones featured very different characters, and they were in France and Mexico, respectively. Those stories didn't work—mainly because I couldn't get past the abstract idea of "cross-cultural encounter" to produce real characters and a grounded story—but they taught me things I needed to know in order to write "Cowboy." Or take "The Shy Birds of Hope," which is interested in professional envy—wanting to be a standout person in one's field. The story in the book focuses on a naturalist who stars in a PBS nature show and wants to reach more people, but my first attempt at the material starred an Olympic archer who always gets silver medals. (That angle was so clearly terrible that I never even finished the first draft.) I could give other examples; behind the stories in this book are hidden a number of false starts. Or maybe they were true starts—true in a more complicated way.
And then there's the kind of revision where you write a story that is basically on target, but you don't get every bit of what you were after, so you write a second on-target story, and maybe others beyond that, and you keep all of them. As the painter Philip Pearlman said, "An artist needs a problem he can't solve. The lucky ones get into a problem that is unsolvable, so they keep going and there's a growth, evolution." When you're in this situation, you might produce a number of stories you can stand behind, all of which are after similar things. The main concept behind The Guy We Didn't Invite to the Orgy and other stories—the common feeling of not quite fitting in—drives the title story, which is all about being left out, but it also drives "Everyone Around Me," "The Four Seasons Club," "Out of Grapes," "Comedy of Errors," and, in fact, "Cowboy" and "The Shy Birds of Hope," among others. The social world is a kind of unsolvable problem, I think, and yes—I feel lucky to be engaged with it.
These days I think I bat about .500 when it comes to short stories; about half are keepers. Some of the failed stories I write are complete dead-ends; they don't go anywhere and they're not really practice for other stories (except in the loosest possible sense that all writing is practice for all future writing). But others are steps on the way. And that means that they're not failed stories at all.