News Ticker

MIND MELD: Scientific Accuracy in Stories

Science fiction would be nothing without the science. Who doesn’t like reading about new or interesting ideas inside of a story? But should SF authors know their stuff when it comes to the science behind the stories? To that end, our question this week:

Q: Do science fiction authors have an obligation to be scientifically accurate with their stories? Is there a minimum level of accuracy an author should adhere to?
Alastair Reynolds
Alastair Reynolds is a science fiction writer and former scientist. He lives in Wales. His latest novel is the far-future House of Suns.

No, science fiction authors don’t have any obligation to be scientifically accurate – up to a point. If we insisted on absolute scientific verisimilitude, then – at a stroke – we’d take out some of the best stories and books the field has produced – Dick, Sturgeon, Cordwainer Smith – pretty much the entire “New Wave” and a lot that’s come since. But there have to be limits, and I think (since most written SF does at least give lip-service to scientific accuracy) this is best illustrated in relation to media SF. Star Trek, for all its undoubted faults, always exhibited a basic grasp of the scale and structure of the universe. The writers and producers, despite their reliance on “creatures of pure energy”, the “particle of the week”, and other such hokum, did at least understand that planets went around stars, that stars were an inconveniently long way apart, that the galaxy was composed of billions of such stars, and all the other galaxies were sufficiently far away that they may as well not exist. It was understood that warp drive was a necessary prerequisite for interstellar journeys, whereas impulse drive sufficed for tootling around the solar system. Compare and contrast this creditable stab at realism with the lamentable Space:1999 (which I nonetheless loved with an unbridled passion when I was 9) and there’s little or no sense of the writers having any grasp of the rhetoric of scale. Planets, suns, galaxies, etc, all seem to be essentially interchangeable entities. No known physics could ever account for the speed with with the runaway moon zipped from planet to planet, while remaining close enough to any given planet to facilitate back-and-forth travel by rocket for an entire episode’s duration. Star Trek‘s writers understood that the Enterprise would require inertial dampeners if its crew weren’t to be squashed by the immense accelerations associated with star travel. No such consideration was ever part of the thinking behind Space:1999, in which the entire moon was blasted out of Earth orbit with no repercussions beyond a few broken items of furniture. In my view, Space:1999 is beyond the pale and can’t really be enjoyed on any level except as dreamlike fantasy, filled with science fictional props that nonetheless don’t fit together in any coherent fashion. But the Eagle transporters did look way cool.

There’s a wider point, though, this is this: why would anyone not be sufficiently enthralled and interested in science to want to get it right? Science is an inherently fascinating and rich human enterprise. We would rightly scorn any writer who professed to a disinterest in the facts of history, or psychology. But science is too often seen as some kind of optional intellectual add-on, a bit like having an interest in early music or Swedish cinema.. That’s not to say that all SF should be scrupulously accurate, all-dots-crossed, my slide-rule’s bigger than yours Hard SF – it would be boring if that were the case. But I don’t think the injection of a tiny amount of real-world science has ever hurt a story, and it’s definitely helped some of mine.

Chris Dolley
Chris Dolley is the author of Resonance and Shift, a pioneer computer games designer, and the man who convinced the UK media that Cornwall had risen up and declared independence. His website is:

I think this all depends on the type of story and the definition of scientifically accurate. Our knowledge of the universe is incomplete and evolving. Some stories, which a century ago would have had no trouble being accepted by the mundane SF magazines of the time would be tossed into the bin today. And anyone pushing a 21st century hard SF story in medieval Europe would have a string of irate one-star reviews on Not to mention a visit from the Inquisition.

Theories come and go – that’s the nature of scientific progress. Natural selection weeds out the craziest and… (I wanted to say throws them away but I think there may be a case for suggesting that natural selection has a soft spot for the really weird counter-intuitive theories like quantum mechanics) Whichever, the result is that at any one time there are competing theories and subjective interpretation of data.

Which means that anyone writing a story today that deals with global warming is likely to annoy someone, who’ll probably cite poorly understood science – i.e. not their version of it – as the reason the book had to be thrown against the wall and ceremonially burned (or composted depending on your carbon viewpoint:)

Then there are the sub-genres within SF. I’d imagine that hard SF readers would be less forgiving of scientific elasticity than their softer counterparts. As for myself, I have a high threshold of belief suspension and am quite willing to read a book where the author maintains the world is flat – as long as he/she writes a compelling and internally consistent story.

Which is probably the reason I write at the science fantasy end of the SF spectrum. I like taking fringe theories and running with them. In the same way an alternate history author bends history to explore an interesting ‘what if’ scenario, I like to bend science. Does that make my science inaccurate? From a mainstream perspective, yes. But I prefer to label it unproven.

David Louis Edelman
David Louis Edelman’s first novel, INFOQUAKE (SF Signal review), was called “the love child of Donald Trump and Vernor Vinge” by Barnes & Noble Explorations, and later named Barnes & Noble’s Top Science Fiction Book of 2006. The novel was also nominated for the John W. Campbell Award for Best Novel. Edelman’s next novel, MULTIREAL, will be released by Pyr in July 2008.

Science fiction authors do not have an obligation to be scientifically accurate in their stories. Why? Because scientific accuracy isn’t the only goal of science fiction. “Science” is only half of the name, after all. Sometimes the whole point of a science fiction story is to explore the impossible, or to conduct a wild-n-crazy thought experiment, or to examine the very premise of scientific accuracy in the first place. There’s no reason to straightjacket stories by insisting that they conform to the accepted scientific norms of the day.

Besides which — and you’re going to roll your eyes at this one — what is scientific accuracy anyway? Not only do scientists freely admit they don’t know everything, but they often speculate much the way that science fiction does. These days, talk of the metaverse, time travel, and alternate realities isn’t just geekspeak at an SF convention; it’s freely bandied about in respected scientific journals. I read a treatise not too long ago by a guy who put forth a rather convincing argument that we’re actually living in the Matrix.

And let’s not forget that the bar moves quickly. While Mundane SF authors are busy manifestoing about how FTL (faster than light) travel is a ludicrous impossibility, real-life physicists are theorizing about ways in which an Alcubierre drive could theoretically warp space and allow FTL travel after all. No matter how wild us SF geeks get in our speculation, we’re never going to approach what the physics geeks are already imagining.

So we have no obligation to be true to some static definition of scientific accuracy. That being said… black people have no obligation to avoid naming their kids Amos and Andy either. But it doesn’t hurt to have a little sensitivity about it, if only to avoid bringing up offensive and inaccurate stereotypes from the past. Science fiction already has enough of a stigma as it is. It couldn’t hurt for us to be mindful of perceptions in the service of bringing the genre to a wider audience.

Marianne de Pierres
Marianne de Pierres is the author of the award-nominated Parrish Plessis and Sentients of Orion series. Her short fiction has appeared in various anthologies and magazines. She is currently working on a SF film project for Sydney-based, Enchanter Productions, called Stalking Daylight, and her new novel Mirror Space. This year will see the release of Chaos Space and her shared world fantasy novel for children, Citrine.

I don’t think that authors have any ‘obligation’ at all to be scientifically accurate in their novels. However, I do believe that writers are foolish to ignore the need for internal logic in their stories; whether that be with regard to science or ‘magic’ or any combination of the two. Science fiction readers are intelligent and unforgiving but they also want to be awed. If you (the writer) can stretch the boundaries/limitations of current scientific understanding in a logical manner, they’re likely to accept and enjoy it as well. If you (the writer) mistreat common scientific understanding e.g. your earth is flat, then you need to have a damn good reason within your story to support it.

Alexis Glynn Latner
Alexis Glynn Latner‘s science fiction novel Hurricane Moon was published by Pyr in 2007. Twenty-three of her novelettes and short stories have been or will be published in science fiction magazines, especially Analog, and horror and mystery anthologies. Her short story “Kindred” in the anthology Bending the Landscape: Horror won the 2002 Spectrum Award for best short fiction. From 2004 until 2007 she was the South/Central Regional Director of the Science Fiction and Fantasy Writers of America (SFWA). She also does editing, teaches and coaches creative writing, and works in the Rice University Library. For fun and real-life adventure, she flies sailplanes and rides in small airplanes as often as possible.

If your readers include scientists, you better get the known science right. You can graft imaginary science onto the real thing. You can posit faster-than-light travel, residing in some unknown gap in the laws of physics, but you can’t just arbitrarily break the cosmic speed limit. Or if you do, you’re writing fantasy, and the whole story should be consistent with that. I’m not aware of any readership that favors a crazy quilt of plausible science and random fantasy.

Accurate science means adequate research and expert review. Fortunately, it’s easy to network around to scientists who feel honored to vet a manuscript. They’d rather catch scientific errors before publication than feel put out afterward. I think it’s almost impossible for a writer to accurately render the facts and the feel of any field not their own; but due diligence followed by expert review goes a long, long way. That, and venturing no deeper into technicality than the story warrants. I once heard a biologist lament a book in which the writer took pains to describe an application of x-ray crystallography and got it wrong. The biologist would have been quite happy with a passing reference to a cool machine, but was offended by the lengthy and implausible description of same.

So if you go into details, also go to experts, and get it right. There’s a lot to get right: terminology, theories, and the scientific method; the mindset of scientists, their work habits, and the places where they work; and the particular worldview of a scientist in a given field. There are distinct differences in the outlooks of cosmologists, chemists, geologists, and computer scientists, not to mention ecologists, biomedical researchers, and astrobiologists!

The other day at a party, I heard a charming little statement from someone who writes patents for applications of nanoscale science. Apparently in writing a patent it’s OK to cite other literature for descriptions of lab techniques or equipment in general use, but everything unique to what’s being patented must be completely described. Said the patent writer: “I write science fiction!” Describing science on the cutting edge in enough detail to substantiate a patent is, according to my informant, incredibly tedious work. Those of us who write science fiction stories and novels might do well to enjoy a bar that isn’t set that high, and go play with ideas, but get the science right enough for our scientist readers to have fun too.

Nancy Kress
Nancy Kress is the author of 21 books of SF, fantasy, and writing advice. She has three more books appearing in 2008, a collection of short stories and two novels. Her fiction has won three Nebulas, a Hugo, a Sturgeon, and the John W. Campbell Memorial Award.

I think there are two (at least) kind of SF stories. Those that are trying to create a believable alternate future do need to conform to accurate science; an example is anything by, say, Jack McDevitt. The second type uses the tropes of SF not to create a believable future, but as metaphors to say something about the present. One example would be Neil Gaiman’s story “How To Talk to Girls at Parties.” Nobody bellieves these alien girls are making real contact with Earth in this way, because Neil is after something else entirely. Both kinds of SF are legitimate, and I’ve wriiten both. The mistake is to judge one as the other.

Karl Schroeder
Having wracked his brains to be innovative in the novels Ventus, Permanence, and Lady of Mazes, Karl Schroeder decided to relax for a while and write pirate stories, starting with last year’s Sun of Suns and Queen of Candesce. Of course, these novels are pirate stories set in a world without gravity — but hey, swashes are still buckled, swords unsheathed, and boarding parties formed in the far-future world of Virga. He’s currently writing the fourth book of the Virga series (no, it’s not a trilogy) and thinking about how to hammer science fiction into some new shapes based on current research into cognitive science. When he occasionally pokes his head out of the trenches, he blogs about this stuff at

Let’s turn that question around: do scientists have an obligation to be scientifically accurate with their theories?

If by that we mean ‘true to the picture of the world we have so far’, then the answer is clearly no: scientists have an obligation to attempt to falsify that current picture (they hope to fail). They are obligated to be inaccurate by definition because science progresses by finding the exceptions to our common understanding of things; the inaccuracies in the current ‘accurate scientific picture’.

If scientists are obligated to look for holes in the ‘scientifically accurate’ picture of the world, would it make sense for science fiction writers to be obligated to uphold that picture?

I’m just askin’.

Elizabeth Bear
Elizabeth Bear was born on the same day as Frodo and Bilbo Baggins, but in a different year. This, coupled with a childhood tendency to read the dictionary for fun, led her inevitably to penury, intransigence, the mispronunciation of common English words, and the writing of speculative fiction.

None whatsoever. With one notable exception, which is to say, when writing rigorous quote unquote hard science fiction. I do think the SF writer has an obligation to know which rules she’s breaking, and break them for a purpose, as an author writing historical fiction should alter history with intent rather than from ignorance.

I think of it like Stephen King rearranging the geography of Maine to suit himself.

Fiction is, after all, fictional.

Adam Roberts
Adam Roberts was born two-thirds of the way through the last century; he presently lives a little way west of London, England, with a beautiful wife and two small children. He is a writer with a day-job (professor at Royal Holloway, University of London). The first of these two employments has resulted in eight published sf novels, the most recent being Splinter (Solaris 2007) and Land of the Headless (Victor Gollancz 2007). The second of these has occasioned such critical studies as The Palgrave History of Science Fiction (2006).

Accurate to what?

I am not a scientific relativist. I don’t believe that the laws of gravity or the conservation of momentum vary depending upon one’s cultural context, since clearly they don’t. Nevertheless, I don’t agree that ‘science’ is a single, monolithic or ‘we all agree upon this’ sort of business. Science is a complex, multifacet praxis; and the way we do science (or the way we think about doing science) is going radically to shape the answer to this question.

My understanding of ‘science’ has been profoundly influenced by the American philosopher Paul Feyerabend (1924-94). His book Against Method (1975) is a powerful polemic against ‘method’ in science. The best way to do science, says Feyerabend, is anarchically; anarchism, whilst perhaps not the most attractive political philosophy, is certainly excellent medicine for the ‘philosophy of science‘ he says. Scientific rules limit possible advances in science: ‘the only principle that does not inhibit progress is: ‘anything goes‘;. Feyerabend proposes a free-for-all proliferation of scientific theories, even though some of these theories will be kooky, daft or unpalatable. Howsoever odd these theories get, Feyerabend is sure that in their interaction better and better models will emerge, better and better science will be practised. The alternative, he says, is to propose a uniformity, a situation in which the powers-that-be by force compel consensus. Feyerabend argues that ‘proliferation of theories is beneficial for science, while uniformity impairs its critical power. Uniformity also endangers the free development of the individual.’ So for example, conventional science was not apprised of the environmental dangers of technological advance; awareness of such issues was raised by groups outside science, ‘Green’ political advocates, New Age enthusiasts and cranks of all sorts. And yet such figures have been vital in broadening useful debate on global warming, the environmental impact of technology, carbon-economy; all things that ‘Science’ now takes seriously. Feyerabend says:

Non-scientific procedures cannot be pushed aside by argument. To say: ‘the procedure you used is non-scientific, therefore we cannot trust your results and cannot give you money for research’ assumes that ‘science’ is successful and that it is successful because it uses uniform procedures. The first part of this assertion is not true, if by ‘science’ we mean things done by scientists – there are lots of failures also. The second part – that successes are due to uniform procedures – is not true because there are no such procedures. Scientists are like architects who build buildings of different sizes and different shapes and who can be judged only after the event, ie after they have finished their structure. It may stand up, it may fall down, nobody knows. [Feyerabend, p.2]

Of course Against Method is polemic rather than manifesto for change in science, and it is perhaps hard to see how his ideas might be put into practice in real terms. Grant awarding bodies, after all, do need some criteria to judge who gets research money and who doesn’t, there being many more applications than money to fund them. And yet it is the case that there does exist a space where the sort of ‘science’ Feyerabend is proposing already takes place; where brilliantly unorthodox thinkers throw ideas around regardless of how strange they seem at first; in which experiments are conducted, and blue-sky research undertaken. This space is called science fiction.

So, as against a rigid understanding of scientific ‘truth’ I’m putting forward an anarchical Feyerabendian sense of the term (with correlatives of ‘imaginative intellectual play’ and ‘extrapolation’). Of course, many SF writers and fans take a particular pleasure in the correctness of the science of science fiction, ‘correct’ here being understood as ‘not transgressing the laws of science as they are presently understood’. In Deconstructing the Starships Gwyneth Jones’s asks: ‘does it matter if the science is sound? The fantasy fanciers will say no, the sf faithful will say yes’. She goes on to point out that Larry Niven’s Hugo- and Nebula-Award-winning Ringworld (1970), ‘one of the great, classic “engineering feat” sf novels, reached print in the first instance with terrible mistakes in its science’, and that Niven, ‘as free as any sf novelist alive from moral qualms about social verisimilitude or cultural relativity, acquiesced to the helpful advice he received from Dyson Sphere buffs, and obediently corrected his fantasy for later editions.’

The shibboleth here is consistency, and one problem with its application is that fans tend to overlook substantive transgressions of scientific orthodoxy (spacecraft that can travel faster than light) whilst becoming agitated about minor features (the mechanism by which Niven’s ‘ringworld’, a massive ring of habitable land circling a star, is kept precisely in its orbit). This inconsistency in applying—precisely—criteria of consistency reveals an ideological ground; for only an ideological belief in science as ‘truth’ can sanction the sort of misprision necessarily perpetuated by this sort of analysis. Another example: Robert Lambourne, Michael Shallis and Michael Shortland’s Close Encounters? Science and Science Fiction (1990) analyse various SF tropes. Space habitats or spacecraft that are spun to give the illusion of gravity in a free-fall environment is a popular recourse of the SF text, in part because such centrifugal environments avoid the need for the ‘pseudo-science’ artificial gravity. Lambourne et al discuss the way the coriolis effect, created by the constant rotation, would determine life inside such an environment.

In the short story “Small World” (1978), by Bob Shaw, for example, a projectile is described as travelling across a cylindrical space habitat along an S-shaped trajectory. In fact, the reversal of the Coriolis force after the projectile passes the midpoint of its course and starts it descent, means that the path is C-shaped when viewed from the drum, as shown in figure 5(b) [Lambourne et al., p.55]

The category error here is the ‘in fact’. A story is not ‘fact’; nor does fictional entry into one or other discourse of science render it so. Application of conventional scientific orthodoxy as a criterion of judgment for an aesthetic object is limiting even when applied with absolute consistency; and when applied inconsistently, as it often is (swallowing the camel of faster-than-light travel but straining at the gnat of, for instance, S-shaped ballistic trajectories inside spinning environments) it combines deadness with muddle. Our choice is between a textual universe run along the oppressive lines of a monolithic unitary ‘truth’ of science, or a science fiction that plays anarchically with ‘science’ along the lines Feyerabend suggests. This seems to me no choice at all.

About JP Frantz (2322 Articles)
Has nothing interesting to say so in the interest of time, will get on with not saying it.

6 Comments on MIND MELD: Scientific Accuracy in Stories

  1. Adam Adam Adam… Feyerabend? did it have to be Feyerabend? It’s okay, I’m not angry I’m just disappointed. I’m disappointed that you’ve let yourself down. You’ve let yourself down, you’ve let your family down, you’ve let the whole school down.

    I think you should go back to your room and think about what you’ve done and whether this is the kind of person you really want to be.

    So terribly disappointed.

  2. Ah, Paul Feyerabend… the great romance of my youth.

    He’s still good, but I’ve moved on past sociological interpretations of science these days; my favourite model now holds that science is a form of distributed cognition (see Edwin Hutchins’ Cognition in the Wild) and the fact that key steps in its cognitive processes are done in and ‘by’ the physical world itself is what accounts for science’s unique ability to ascertain ‘truth.’ (Quotes in deference to those postmodern holdouts who might otherwise quibble me to death.)

  3. As a scientist who reads a lot of science fiction, I don’t have a problem with liberties taken with science – in fact, when done well it can improve things significantly.

    Of course, when the science discussed ends up in my area of expertise (geology) it’s harder to dismiss those liberties. They often annoy the hell out of me and I find myself focusing on them far too much. However, that’s not an issue with the author, but a consequence of my specific place as a reader.

    Basically everyone seems to agree that fiction is expected in science fiction.

  4. Jonathan M: you’re harshing my buzz, man.

  5. As a mere reader, and a “hard science fiction” fan (scare quotes because I’m not talking Analog-style hard sf so much as Al Reynolds/Paul McAuley-style), I agree with, er, most of the voices here. I do like the parenthetically-mentioned Mr Reynolds’s view. (And I too am a little dismayed at Mr Roberts’ use of Feyerabend, buzz or no buzz :))

    What is important to me as a reader is that that when science is used in a story, I get the feeling the author has a decent idea of when they’re being accurate and when they’re taking liberties. Making egregious mistakes to no real end loses the respect of the informed reader – at least in the reader’s areas of expertise.

    See the above comment from a geologist, who is less forgiving of errors in geology. While I have a background in science, I am also a professional musician, and I am troubled by unconvincing use of the musical world in stories (indeed, a little bugbear of mine is the use of “crescendo” to mean “climax”. It’s Italian for “growing”, and means a gradual increase in intensity folks, m’kay?)

    It’s telling that Alastair Reynolds actually contacted me while writing Century Rain to get some advice on the plausibility of his use of a double bass case in the novel!

    Of course there are plenty of areas where the facts are hard to discern, due to some controversy or other, or due to the difficulty of the field in question. Even then, it’s possible for a good writer to hint at this through subleties of characterisation and use of narrative voice (or hey, through a good old fashioned infodump if you like).

    And this brings us to another one of those sf bugbears: convincing characterisation. Is it important that your characters have individual voices, convince as representatives of their gender/ethnicity/socioeconomic status/sexual preference?

    Well of course it is, but depending on the genre, intentions of the author, target audience and so on, you can play faster or looser with any such aspects of story construction. I’d prefer to have compelling and subtly-drawn characters, but in a short work of hard sf or a fun thriller one might not mind too much.

    It’s up to the reader to decide how much veridicality they want out of the various facets of the stories they read. In the end, the author has to be careful to keep the reader engrossed and believing throughout – whether it’s a weird and wonderful piece of experimental prose, a slipstream piece about Kabbalistic AIs, or a far future extrapolated paean to technological progress… Sloppy details that don’t ring true are going to jolt your reader out of the story.

    And it seems to me if the reader stops believing in you, you’re in trouble.

  6. patrick // May 27, 2008 at 4:52 pm //

    Some very interesting commentary here. And mostly balanced responses. My simple, and perhaps somewhat vague, answer to the question here is:

    Obligations are for those who feel they need them. However, I must point out the difference between sci-fi (or, as Dan Simmmons calls it, skiffy), and SF. (This latter was implicit in some of the commentary above – for example, Reynolds’.)

    And, I’ll one-up Karl. (Not intentionally, of course – as I’m a leisurist, and am not emotionally prone to promoting competition – but incidentally.) Assuming cognition at all is close to admitting free will, and both share an anthropomorphic bed. A, what I call, funxional approach is to intuit beyond a human frame. A start is to transcend culture. Another step is to transcend gender. Note that the use of elements of these things does not contradict such transcendence, as they surely have artistic utility – given context.

    Or, in perhaps more poetic words, emotionally vibrant-yet-unattached. Of course, where does this leave the writing of stories?

Comments are closed.

%d bloggers like this: