Wednesday, August 24, 2005

Writing Speculative Fiction

I just finished reading (re-reading, actually) Robert Heinlein's classic essay, "On the Writing of Speculative Fiction." If you haven't read this short (six pages in the edition I have) work, you owe it to yourself to do so. Many authors (including yours truly) have said pretty much everything that is contained in this essay before, but no one has said it more succinctly (and while I don't know this for a fact, I suspect Heinlein also said it first).

Heinlein makes a number of points worth repeating. First, he defines two types of science fiction stories, the character story and the gadget story. (Long-time readers of this column will remember that I tend to follow Orson Scott Card in defining two other types of stories, the milieu story and the event story.) Heinlein says he only writes character stories (which is more or less true), so he restricts his focus to that branch of the SF tree. He claims that there are only three plots in SF: boy-meets-girl (and all the variations thereof), the Little Tailor (a small guy who makes it big, or vice versa), and The Man Who Learned Better. I laughed when I read this, thinking, "Nah, no way -- there's more to it than that," but for the life of me I can't think of a counter-example. The novel that I've just started is pretty solidly in the "Little Tailor" camp, as are most of my other tales (I tend to write about the underdog). A lot of years have passed since Heinlein wrote this essay, so I'd really be interested to hear if someone has come with a new plot!

Heinlein gives five rules specifically related to writing science fiction. I happen to strongly agree with all of them, with qualifications (you can read my earlier discussion on the topic here):

  1. The conditions must be different from here-and-now.
  2. The new conditions must be essential to the story (this is, in my opinion, the key requirement for something to be considered science fiction)
  3. The problem (the plot) must be a human problem.
  4. The problem must be created by or dramatically affected by the new conditions.
  5. No established fact can be violated, and any new theory brought in to replace an established one must be fully explained and justified.

I would be careful with this last one. I think no other dictum has led to as much boring infodump in science fiction as this one. Yes, you need to completely work out all the implications of changes to known theories -- and your changes must be just as plausible as the original theories. But you do not, however, have to give your reader all of these details. Instead, find ways to work the consequences of your theory (rather than the theory itself) into your story. There's just no excuse for a physics lesson right in the middle of all the action.

This essay also contains Heinlein's famous five rules for fiction writing in general. I'll quote them here, the parentheses are my own comments:

  1. You must write. (Preach on!)
  2. You must finish what you start. (Amen!)
  3. You must refrain from rewriting except to editorial order (ah, no -- but we'll get back to that)
  4. You must put it on the market. (You may be a writer, but you are not a professional writer until you submit your work to be published -- whether or not it is accepted.)
  5. You must keep it on the market until it is sold. (You may be a professional writer, but you are not a successful professional writer until you have made a sale.)

I disagree with #3, though I do understand where it comes from. Heinlein was a writer for his day. At that time, anything and everything remotely related to science fiction was selling and selling well. Editors established personal relationships with writers, and quite a few mediocre writers were nurtured by these editors until they became outstanding authors. When Heinlein talks about "editorial order," I think this is really what he is referring to. Avoid the friends and co-workers who have "ideas" for your story and trust the professionals to whom you are selling the work. Unfortunately, the publishing world just doesn't work that way anymore. Editors no longer have time to give you feedback on how to make your story publishable. As a result, you have to become your own editor -- and that means rewriting and rewriting again until your manuscript is in its absolute best form. Notice I didn't say "perfect form," because a manuscript will never be perfect. If you keep revising forever, you will violate #4 and will not be a professional writer. Get it as good as you can, given your current level of ability, and get it out the door. And when it comes back, send it out again, and again, and again. Eventually, it will find a home.

1 comment:

Steve said...

Thanks for summarizing Heinlein's advice, Keith. It's hard to find on the Web, surprisingly. We linked to at The Morning Mug.