How Much Science in Science Fiction?

How much science do you have to know in order to write science fiction? It’s a simple question with a complicated answer:  A whole lot. Some.  Make it up.  Speculate.

The only wrong answer is “none,” if you’re calling it science fiction.

I’m not a scientist.* And it isn’t necessary to be one in order to write credible science fiction; SF is about ideas and people dealing with those ideas, not just about science.

Do your homework.

In search of answers, I attended a panel about the science in science fiction at the 2016 Philcon.  The panelists were Ian Randal Strock, Mike McPhail, Jane Fancher, John Skylar, Phil Giunta, and David Walton.  All of them, as they were introducing themselves, claimed not to be qualified to be on the panel, so I immediately felt at home.

However, it was clear they covered a spectrum of “not-qualified” and so did their opinions.   Some readers can be very opinionated.  As Strock said,  “Readers of Analog read with a calculator,” which is not necessarily true for, say, Fantasy & Science Fiction magazine.   He suggested one way to get the science right enough for the picky reader is to write for the far future, and if science catches up and passes your story premise, “read it as alternate history.”

According to Jane Fancher, “A lot depends on how important it is to the story.”  I would argue that all of it depends on its importance to the story, but that’s true of any story.

David Walton,  whose bio on his website says he writes “quantum physics thrillers,” said, “There is always going to be someone who knows more than you do,” and said you have to “get the culture of science right.”

As an example of getting the culture of science right, John Skylar mentioned C.J. Cherryh’s Cyteen, and also Gregory Benford and David Brin’s writing.  Like them, you should understand the differences, say, between a pharmaceutical research lab and a defense research lab.

However, Skylar suggested another approach, especially when something just isn’t possible:  He says, “You can’t?  Well, what if you can?” In other words, writers get to pretend.  Also, “What if you did the experiment wrong?” makes for a compelling premise as well.

However, Phil Giunta said he was “not comfortable disagreeing with science.”  He added, “Research is part of your writing life and worth your time.”

All too often, though, as writers are drafting, the science catches up and passes them.  Cell phones are the most compelling case.

Sometimes, however, writers anticipate technology.  Mike McPhail used Larry Niven’s autodoc as an example; in the technology’s first appearance in his books, Niven didn’t explain how it worked, but, as tech caught up, he applied it in his writing.

And how are you supposed to find out all this important information?  Go and ask, they agreed.  Don’t try to get people to do all your work, but read up on your topic and write or call saying, “I’m writing a story about,” with brief, specific questions.  Or you can ask on Reddit, or do an Ask Metafilter.  As long as you are not too demanding, as long as you are courteous about it, people are surprisingly willing to help.

My take:  Nobody quoted Clarke on “sufficiently advanced technology,” but it’s a good rule of thumb.  Don’t over-explain, don’t turn your story into a treatise.  In the future, people will take future technology for granted.  Some writer lost in the mists of time (John Campbell?) said when a character pulls out a ray gun in a science fiction story, she doesn’t stop to say, “This is a ray gun,” and explain how it works.  No.  She shoots the ray gun.  But she should shoot it in such a way that it sounds credible.

Next:  Instruments of Mass Creation

===========

* Actually, I know a great deal about science, and anyone who writes science fiction should be familiar with both of C.P. Snow’s cultures.  I am uniquely qualified to be unqualified in both bultures; though I taught English for 12 years and science for 10, my degree is in fine art.
Here’s the thing, though:  My grandfather, two uncles, and my mother were hard scientists.  My mother was getting her Ph.D. when I was 12 and I hung out in her lab.  I taught elementary school science for ten years.  I know a fair amount about linguistics, anthropology, sociology, mathematics, chemistry, and physics, among other topics.  I have read Kuhn’s original paper on paradigm shifts, and I can discuss Popper, Darwin, and Mendeleev.  I worked for a center for the history of chemistry.  My own doctoral work incorporated a wide array of disciplines.  I have done research, and I got an A in statistics.  And I read a whole lot of science journalism and nonfiction of many kinds.  But I am not a scientist. I’m a SF and fantasy writer.
If I was a scientist I would be writing grant applications right now and worrying about funding.  Or I would have retired and be writing science fiction, more probably.
† I have a draft of a novel that was written ten years ago. It will have to be completely rewritten because of cell phones.

10 thoughts on “How Much Science in Science Fiction?

    1. DMT says:

      I would think you’d HAVE to have good science (and “Otherwise, why read?” is a valid question then) if the premise is about “the machine that could solve the world’s energy problems” and your story is set close to the present day, as OIL AND WATER seems to be. The imperative is to make the story believable somehow, whether it’s because the research is good or because the motivation, characters, and setting seem believable.

  1. Dianna says:

    I think it also depends a lot on what you’re writing. You can get away with a LOT of weird non-science in TV and movies that you’d never get away with in a book because people actually see it and therefore it’s easier to believe.

    1. DMT says:

      That’s for sure! Inception was an example of that for me. I loved the movie but the premise was hocus-pocus. Or is it just that we accept different conventions in movies than in books? There sure is a lot of noise out there in vacuum when spaceships fight on screen.

  2. Clare Deming says:

    Yes, agreed! The same can be said for medicine and injuries in fiction. I think that finding a critique group with members of varied backgrounds and experiences can help. You may not even realize that something in your story needed to be researched, or that maybe you misinterpreted the research. It’s good to have someone else point out where it wasn’t believable.

    1. DMT says:

      I’ve been thinking about what you say, and I bet inaccuracies of medicine and injury in fiction set off your alarm bells big time, since that’s your area of expertise. The same happens to me when I read naive descriptions of classrooms (or, for that matter, the rants about education that people seem to default to when complaining about politics). When I really know something about a subject, it is often painful to read people’s descriptions. But I don’t know if fiction writers always have to write for the expert – we just have to make it credible and not get things completely wrong.

      1. Clare Deming says:

        Yes – it doesn’t have to be perfectly accurate. This is fiction, after all! I just groan a bit when medical things are way off. Animals too – horses are often screwed up in fantasy, for example. I mean, they never colic or founder. Ever.

        1. DMT says:

          Ha! That’s too wonderful. And it’s true. They are sort of meat vehicles in many novels. I often wonder how they can go so far. But then I wonder that about the humans in many of those cases.

Leave a Reply

Your email address will not be published. Required fields are marked *