How much science knowledge do you need to write science fiction?

Edward Willett @ 16-06-2008

Tom Swift Cover On her blog, author Jo Walton laments that:

I can’t write science fiction because I know both too much and not enough science.

I know too much to spout total crap and not care, and I don’t know enough to inherently get it right. So I can write it and be sort of right and I need to get it checked.

(Via io9.)

But getting it checked, she goes on to complain, slows her down so much that she can lose momentum and be unable to write the story at all:

The way I write, I inclue as I go along and plot develops as I go along and background develops out of that, and my understanding of the world develops (even if lots of it doesn’t end up on the page) and if half of what I think turns out to be wrong then it just gets to the point where it isn’t worth doing in the first place. The people who know science suggest alternatives that totally screw up what I wanted to do and why I wanted to do it, and I lose all confidence in it and decide I should stick to stuff I understand.

She then gives a specific example.

I know where she’s coming from: I got held up quite a bit on my most recent novel, Marseguro, while I tracked down the information I needed to ensure that my spaceship’s habitat ring rotated at the correct speed, given its diameter, to generate something approaching 1 G in the outermost layer–and that at the central core my characters could believably make the transition from non-rotating section to rotating section without getting their arms ripped off.

Non-SF writers never have to worry about stuff like that.

So: if you’re a writer, how much time do you devote to getting the science right, and if you’re a reader, how much accuracy do you demand? (Movies, of course, are a whole different kettle of fish where even non-SF films never get the physics right.)

(Image: Wikimedia Commons.)

[tags]science fiction, books, writing, novels[/tags]

Be Sociable, Share!

11 Responses to “How much science knowledge do you need to write science fiction?”

  1. Ben says:

    You’ve wrongly changed her “inclue” to “include” – incluing means giving the reader small clues about the setting, as opposed to infodumping.

  2. Denni says:

    Oh, I’m in exactly the same shoes. I’ll spend months on background research, and I’m mortified if I get it wrong. Consequently it will take me years to finish my current novel (‘Tree of Life’). It can’t be helped.

    On the other hand, while I want the biology to be at least plausible, I’m quite happy to insert some vague entities such as nanites and Star-Trek-type replicators 😉

  3. Edward Willett says:


    Oops, you’re right. I’ll fix that. Thanks!

  4. JB Dryden says:

    If you’re writing hard science fiction, then I think you need to spend the time to make the science in the story plausible, if not practical. If you’re writing soft science fiction, then I think there’s a bit more leniency as to the authenticity of your facts. Suspending one’s disbelief is part of the process of speculative fiction, but I do agree that at some point, the science that underlies a story (if that’s the point of the story) needs to be believable – even if it’s only for the world that the author has created for his/her reader.

    As an editor, I look for stories that can either make me believe that the facts presented are true, or I look for stories that deal with the science in a way that makes it less a part of the story than the characters. In both cases, I think it’s the duty of the author to present the story in their vision and make me believe in that vision. It really has little to do with how much research they put into it. When you start throwing in quantum physics, I’ll honestly get bored before it gets interesting.

  5. SantaBJ says:

    I’ve never really had this problem, I feel confident enough in my understanding of science that I don’t worry about my writing getting the science wrong. When I *am* uncertain of something, I get it checked by someone whom I know can answer – if the answer isn’t quite in line with what I had in mind I don’t get discouraged like Ms. Walton, however. I just think of it as a new challenge. To me, it’s like “alright – my characters, my story, they have to deal with the science of this. Why shouldn’t I?”. If I don’t get it right, or at least right within the pseudoscientific confines of the story, the story falls apart to me; if my characters can disregard reality to get the job done, what point is there in writing the story at all? Might as well just skip to the end.

    JB Dryden makes a good point when he mentions “hard” science fiction. If you want to write scifi, and don’t feel fully confident in your knowledge of science, then what you should do is lower the bar a little and define your own parameters with regards to the science of the story building on what you know of actual science.

  6. Tom Marcinko says:

    Thanks for the Dalek/Xmas tree cover of the Tom Swift book. I remember it fondly.

  7. Tom Marcinko says:

    I think this is a great question and one that can be pondered to infinity. Obviously it has a lot to do with a particular writer’s process.

    James Blish was quoted as saying that all the science he ever needed he found in a bottle of Scotch.

    Nobody has yet accused me of being a hard-science writer, but I do try to keep up with Science Times, Science News, blogs like Science Daily, and programming on the Science Channel. Part of the task, I think, of deciding what’s plausible is to get a sense of what people know or think about a subject. Plausibility is about consensus, now that I think of it…

  8. Will Hindmarch says:

    As a writer, my confidence gets rattled even about things that are unverifiable — like whether or not my own characters’ decisions make emotional sense. I worry too much about potential readers calling bullshit on any detail, even in obviously fantastical worlds. It’s an absurdity which I suspect all writers have to outgrow, and which many only outgrow so much. But science is just one flavor of the spectrum, isn’t it? Historical novels and militaria and every other sub-genre have their details that a reader can use to disbelieve.

  9. Michael says:

    I don’t have much experience as an SF writer yet, but frankly, I don’t worry about it. I suppose in a sense that makes me a fantasy writer — or really, it makes me a cyberpunk writer more than anything. As a programmer, I create my own reality anyway. And I like the headlong feeling of rush into the future — if my characters are confused about the specifics of what they see, why shouldn’t the reader be equally confused?

  10. Vague says:

    I wouldn’t want to get it wrong, so I’ll probably never write anything remotely like hard SF.

  11. Jeff Fourmyle says:

    “How much science knowledge do you need to write science fiction?” (a) Enough to satisfy your audience. If you’re writing for an audience who thinks like Nebula Award judges, not much. If your stories are set in an environment beyond Clarke’s Limit (science indistinguishable from magic), not much. If you’re writing for Greg Egan’s or Alastair Reynolds’ audience, you better have it nailed.