The Craft of Writing: Credibility in Science Fiction
So you’re writing science fiction. How do you make your future tech seem credible?For far-future science fiction? You really can’t. You can paint in broad strokes what will likely be possible (fusion power) and what will almost certainly never be (faster than light travel). You can take care to learn enough physics that you’re not routinely violating the laws of physics.
But honestly, you’re going to get it wrong. It’s not just that you can’t predict future technology—more important is you can’t predict the ways people will use it, or the effect it has on society.
Some sci-fi writers in the 70s predicted home computers and instantaneous pocket communicators. Nobody predicted Facebook. Some predicted instantaneous worldwide data storage accessible to everyone. Nobody predicted that rather than being used for knowledge and education, it would be used for conspiracy theories about political parties running sex slave rings from the basement of a pizza shop that doesn’t have a basement.
As William Gibson said, “the street finds its own uses for things.”
That’s the part you can’t predict—-you might be able to extrapolate technology, at least in broad shapes, but you aren’t going to be able to see how the future uses it. Your job is to be plausible, which you can do by learning physics and sociology, not to be a crystal ball.
Near future is at least a little easier.
In our post-cyberpunk novel immechanica, Eunice and I tried very hard to make our tech extrapolations of what’s already here or nearly here. A big part of that was talking to people who’ve already worked with the kind of tech we were anticipating.
Truck-mounted Doppler radar to look through walls to spot sniper nests? We talked to Bill Otto, whose suggestions were amazing. (We we’re thinking IR for that originally; he was invaluable in explaining why that wouldn’t work.)
Contact lenses with thin-film displays for augmented reality? Jon Lowry has actually built prototypes, and was able to explain how they work and that yes, the things we wanted in the story were technically possible, if only just barely, by 2054. Brain machine interfaces? We used Neuralink’s most optimistic projections and ran with them.
We also had a very long video conference call with a sociologist and linguist about T-Town, a weird enclave carved out of LA by a former Mexican street gang with, oddly enough, a lot of crossover with Japanese society—which, yes, is a real thing in parts of California.
We extrapolated from modern trends in surveillance, smart autonomous drones, AI, and adversarial input, projected from self-driving cars to try to project trends in future transportation (hypothesis: if full self-driving becomes ubiquitous, a lot fewer people will feel the need to own cars!), that sort of thing.
I’m not saying we got it right, of course. We tried to make what we were doing plausible, something that could happen, not something that will happen.
Which, I think, is the goal.
However you do it, though, whatever world you envision, the tech is not the story. The tech helps set the background. Good stories aren’t about gadgets, they’re about people. Whatever tech you imagine, it’s the way people use the tech that matters.