What is Science Fiction?


Such a simple question, isn’t it?  I mean, we all recognize science fiction when we see it, don’t we?

As it turns out, I had a hard time defining science fiction.  And I found that once I’d formulated a decent definition, I had a hard time writing a convincing argument that didn’t stray all over the place and bog down in inconsequential details.  There are some good reasons for the difficulty, some related to science fiction and others to the concept of genres.

You might wonder why I even ask the question, but for a review site devoted to science fiction movies, it’s a material question.  We all know what movies are, but not everyone can agree about science fiction, as it turns out.  It’s also a fun question to ask and think about, sort of like wondering what the best science fiction films of all time are, or the best novels of all time.

So what makes a good definition?  An ideal definition of science fiction will use objective criteria intrinsic to the work itself.  A good definition should also be short and to the point, which will make it more easily applied.  Finally, a good definition should be specific and simple.  A definition which lacks the two latter qualities will be difficult to apply or give different results when applied by different people—that will always be a problem with the genre of science fiction, I suspect, but we can at least set out to minimize the effect.  The first quality is at least partially a matter of my personal philosophy than anything else; it means that a work should stand or fall on its own merits, rather than by reference to some outside agency.  I do stand by it, however, since it makes so much sense.  We’re going to have problems with that one, by the way.

A good definition of science fiction should also match our common sense, first impression appraisals as much as possible.  If we tend to think of something as science fiction, and application of the definition makes it not science fiction, there’s probably a problem with the definition.  Similarly, if the definition includes a bunch of stuff we don’t recognize as science fiction, it may be too broad.

Difficulties in Defining Science Fiction

Just like a western or a romance or a mystery or an action movie, we all recognize science fiction when we see it.  Potter Stewart said much the same thing about pornography (“I know it when I see it“) in perhaps the most famous statement ever made by a Supreme Court Justice, but he still faced the task of formulating a definition of pornography which lower courts could apply to identify speech unprotected by the First Amendment.  In 1952, the noted editor, writer and critic Damon Knight said that “[science fiction] is what we point to when we say it.”  His statement echoes Stewart’s, and in it he encapsulates the idea that science fiction can be very difficult to define—at least, as he further explained, without bloodshed or conflict.  As it turns out, there has been some disagreement through the years about the definition of science fiction (everyone seems to get upset when a definition gores their ox).

Part of what Knight referred to is that the earliest definitions of science fiction were aspirational in nature:  they focused on what science fiction should be, at least in the eyes of the definers—I’m talking about Hugo Gernsback and John W. Campbell here, who were enormously influential editors (the former essentially created the field while the latter refined it and brought it out of the pulp era and into the “Golden Age”).  An aspirational definition is—at least now, when the field has one hundred or more years of history depending on how you look at it—a little intellectually dishonest, and it also puts the cart before the horse.  A fan or writer or editor should be able to talk about what science fiction is before they begin talking about what it should be.  Any long term fan exposed to these kinds of aspirational definitions may have a hard time ignoring them to get to the meat of the problem, and in any event, such definitions exclude much of what we recognize as science fiction.

Another problem is that science fiction can often have attributes of other genres.  Isaac Asimov, a very famous science fiction writer who also produced a great deal of non-fiction science work as well as mystery stories, advocated the idea that science fiction was a flavor which could be applied to any other genre.  In response to John W. Campbell’s contention that science fiction and mysteries were incompatible, he wrote The Caves of Steel, the first science fiction mystery novel (it also featured a humanoid robot which gave Asimov a Pinnochio character he used to inquire into the human condition, a plot device now familiar to many a television viewer in the form of Star Trek’s Spock and Data, for example).  Other “genre blenders” include Ridley Scott’s Blade Runner, a classic film noir;  Larry Niven‘s Gil the ARM tales, all mystery stories; Anne McCaffrey‘s crystal singer novels (Crystal Singer, Killashandra, and Crystal Line) are certainly romance; Aliens is a pretty clear example of the action-adventure film.  A book or a movie is not any less recognizably science fiction just because it is also recognizably another genre—a view consistent with modern genre theory, which we will address shortly.

Fantasy, in particular, shares this quality of blending genres, but that’s not the only problem that involves fantasy.  The two genres, fantasy and science fiction, are closely related in modern culture.  A lot of science fiction fans are also fantasy fans;  a lot of fantasy writers also write science fiction.  The Hugo Awards, given annually the the works voted the best in their class every year, boast a number of pure fantasy nominees: Randall Garrett’s Too Many Magicians, which is also a murder mystery; Patricia A. McKillip’s Harpist on the Wind, a Celtic-style fantasy; Emma Bull’s Bone Dance, set in a post-apocalyptic world where magic works and subtitled “A Fantasy for Technophiles”; and J. K. Rowling’s justifiably famous Harry Potter and the Prisoner of Azkaban, just to name a few.  Pure fantasy Hugo award winners in the novel category include Neil Gaiman’s American Gods and Rowling’s Harry Potter and the Goblet of Fire.  So it can be difficult to tell these two genres apart, and many fans don’t seem to try.

Finally, sometimes the quality of a work—in the sense of how good something is—gets in the way of identifying it.  Just because something is bad science fiction doesn’t mean it isn’t science fiction.  Many critics and fans get caught up in whether or not the scientific ideas are even remotely plausible as a factor in defining science fiction, when questions of plausibility are probably better examined if you’re talking about whether the science is convincing—a definition which depends on qualities like how convincing the science is cannot be applied in the same way by different people, and moreover looks outside the context of the work to define it.

Genre Theory’s Role in Defining Science Fiction

The field of genre theory is an aspect of literary criticism with roots in math and philosophy, particularly ideas like set theory.  That’s because genre is a classification system which attempts to order literary or cinematic works in groups.  True to form, genre theory is a messy, chaotic field with no clear-cut answers.  The essence of the problem seems to be that fiction expresses the entire range of human experience and imagination, while classification systems do not.

The Greeks, among the first to formulate genres, viewed them as rigid categories (with the exception of certain thinkers like Aristotle, who recognized that the Iliad, for example, was both an epic and a tragedy).  The Romans followed suit, though there is some evidence that not all Romans adhered to a rigid categorization.  As time passed, however, notions of genre expanded.  By the time of the Age of Enlightenment (an 18th century movement which emphasized reason) and the industrial revolution (with the printing press and a a whole new literate class, popular literature, and new forms of work such as the English novel) the older rigid ideas of genre became outmoded.

In a notable essay entitled “The Law of Genre,” French philosopher Jacques Derrida, the founder of deconstructionist thought, put the final nail in the old rigid view of genre’s coffin (Derrida wrote in his native French, and the translator for the version I have read is by Avital Ronell).  Derrida said that individual works “participate” in a genre (as opposed to being members of it), and that the very characteristics which define a genre are effectively uncategorizable.  If you know anything about Derrida and deconstructionist theory, that statement should come as no surprise…Part of what Derrida was getting at is the idea that it is the members of a genre which define the class.  That seems simple enough, but as new works enter the genre they redefine it.

Many modern genre theorists feel that genres, as groupings, are processes rather than classifications.  They also point out that genres are established by critics for specific purposes, and that those purposes often inform the classification process.

Film critics—as opposed to philosophers and literary critics—have taken a much more practical approach when discussing the Western.  Westerns are almost always set in the American west in the 17th century, and they also obey certain conventions and employ certain conceits which form a short-hand method of communication.  The good guy wears a white hat, and the bad guy wears a black hat.  The lone gunslinger who rides into town by himself is probably a really quick draw and a good shot.  If two men in the lone gunslinger or cowboy role are standing facing one another with no one between them, someone is about to get shot.  Cowboys are loners and individualists, while townsfolk are family-oriented.  The conventions of the Western are like codes for the viewer, and assist in expressing the more common themes of the western as well as comprising a shorthand which often moves the plot along.  For the film critic, the conventions and conceits are the critical aspect of the Western.

Dictionary Definitions

The dictionary gives another approach to defining science fiction—the function of a dictionary is to define things, after all.  The American Heritage College Dictionary defines science fiction as “a literary or cinematic genre in which the plot is typically based on speculative scientific discoveries, environmental changes, space travel, or life on other planets.”  The online version of the Merriam-Webster Dictionary states that science fiction is “fiction dealing principally with the impact of actual or imagined science on society or individuals or having a scientific factor as an essential orienting component.”

These two definitions are remarkably similar, which is a good sign—though the aspirational definitions promulgated by Hugo Gernsback and John W. Campbell are also quite similar to one another, and they are dangerously narrow definitions.  Accepting a dictionary definition also robs the definer—that’s you and me—of the chance to think about the issue.  For now, keep the dictionary definition in mind.

Defining Science Fiction

Since the genre of science fiction is a class of individual works, it makes sense to look at recognized works within the category and identify the common characteristics the works share.  The Top Fifty Films list is a good place to start, then, since it’s supposed to be the best of the genre (at least with respect to cinematic science fiction).

A quick perusal of the list reveals a lot of aliens, settings in the future, time travel, space travel, and psionic powers.  But none of these elements are found in all of the films.  The Terminator is set in the present, lacks aliens and space travel, and has no mental powers—but it does have a killer robot sent from the future by an artificial intelligence.  The three Star Wars films are set “a long time ago, in a galaxy far, far away” but each has most of the other elements: aliens, space travel, and the Force as a new take on psionic powers.

These elements have something in common:  be it life forms, places, times, or abilities, they are strange and unusual.  The strange and unusual things—aliens, starships, the Force, killer robots and murderous artificial intelligences— are imaginary.  Importantly, they are discussed or presented within the context of an imagined scientific or technological framework—which is really what differentiates our slowly coalescing definition from fantasy.

The defining characteristic of science fiction, then, is  the presentation of imaginary science or technology in a work of fiction.

This definition invokes an outside standard—science itself—in the definition.  You may recall that right back at the beginning of this essay I suggested that a good definition would be self-contained, and that it would not reference outside standards.  Rather than define science for our purposes—its neither necessary nor desirable in this instance—let’s look at the characteristics of the imaginary science in our films.  What makes the imaginary science science instead of magic?  It is, simply enough, the presentation of the imagined science as science.  If a novel or a film presents something as science, then for the purposes of our definition, it is science.  We’re talking about imaginary work in the first place, and in the second place, such a qualification gets us around the outside standard problem.

You might see the problem, though, and you may have seen it before I tried to finesse my way around it.  How do you recognize the science as imaginary without looking at the state of science and technology as it is now?  The answer is that you can’t.  We’re stuck with an element of the definition—imaginary science—that requires that we recognize real science.  There’s no way around it; we’re simply going to have to live with it.

So…so far, so good.  We have a definition which applies directly to a science fiction candidate without reference to anything outside the film or written work, other than the state of science itself.  The definition is going to distinguish fantasy from science fiction—the fuzzy case, works involving “the mystic sciences” or “the laws of magic” may use a logical framework consistent with science, but they are presenting magic as magic.  There are always going to be instances of a work which blurs the line between fantasy and science fiction, and there’s really nothing we can do about that.  The definition is short and to the point, and it’s specific and simple, at least for most cases.

Does the definition go far enough, though?  Is the mere presence of imagined science or technology enough for a work to qualify as science fiction?  Or must the imagined science be essential to the plot?  On the one hand, an “essential to the plot” requirement will narrow the definition a bit.  On the other hand, application of the “essential” part will call for judgement on the part of the classifier, which is something we wanted to avoid.

Suppose for just a moment that you’ve rented the latest (purely hypothetical—I don’t want an Atlantis on my hands here) film by John Carpenter, Lunacy, featuring a crazed killer loose in Luna City in the year 2125.  It’s a great movie, with lots of scary moments as the madman stalks his victims through the shadowy tunnels and domed parks and technologically sophisticated infrastructure of Luna City, the first city on the Moon.  Is this science fiction?  If we use the broad definition, the mere presence of the imagined technology necessary to build a city on the Moon is enough to qualify it as science fiction.  As you watch the movie, though, you realize that none of that technology is necessary to the plot; the movie could just as easily have been set on Ocrakoke Island or some other isolated location and it would still work.  There are no scenes involving explosive decompression, no mention that the use of firearms in Luna City is banned because of the hazard to the entire city, no spacing of the killer at the end—nothing, in fact, which makes the Luna City setting essential.  Now is it still science fiction?  The broad definition says yes, because the film presents imaginary technology.  The narrow one, in which the imagined technology must be essential to the plot, says no.

We’re going to take the easy way out here, and stick with the broad definition.  Here’s why:  first, requiring that the science be essential to the plot undermines the simplicity of the definition, and calls for judgement on the part of the viewer.  Second, the characteristic of “essential to the plot” looks outside the film itself and it says more about the quality of the work than it does about the kind of the work.  Arguably Luna City is superfluous, at least from the vantage point of the narrative.  Think about the use of the word superfluous in this context, though.  It’s a judgement by the viewer that something in the film or story is unnecessary.  That’s the product of the viewer’s thought process, not something inherent in the film itself.  Carpenter set Lunacy on the Moon for a reason, and the best way to evaluate the film is to simply take it as it is instead of ignoring or disparaging an aspect of the film.  I can also add that, from the vantage point of film as a visual medium, the setting is pretty important.

There are other characteristics of science fiction, but they are either not present in every work of science fiction, or they deal with things outside the works themselves.

For example, early science fiction emphasized  a sense of awe or wonder, which is what I suppose I meant when I talked about the “strange and unusual.”  The most influential early magazines were Gernsback’s Amazing Stories and Campbell’s Astounding Stories, and their titles give voice to this aspect of the field.  To a modern viewer, though, aliens can seem rather blase and ordinary, as can the ideas of faster-than-light travel, time travel, and many of the more treasured—and now common—science fiction conceits.  In other words, awe and wonder are as dependant upon the viewer or reader as they are on the work itself.  Not every work of science fiction evokes the strange; some deliberately do not.

Science fiction is often—and once was very nearly always—positive about the benefits of technology, but again, there are plenty of science fiction stories which present future or imagined technology in a more negative light.  William Gibson, the originator of the cyberpunk sub-genre in the 1980s, emphasizes the negative aspects of the nifty technology he so cleverly invents.  And the dystopian future is a common idea in modern fiction as well.

Science fiction can be eminently plausible, with an attempt to predict the course of future developments.  Again, however, there is plenty of science fiction which doesn’t predict anything at all.  Science fiction nearly always has a “what if?” built into it—that is, the story gives an answer to a question like “what if humanity encounters an alien species?” or “what if a distributed artificial intelligence arises in the internet?”  That’s inherent in the idea of speculation about science and technology.  But not every work attempts to answer that question with an honest-to-goodness prediction of what will actually happen.

Gernsback and Campbell insisted on scientific plausibility in their definitions, and Ray Bradbury once remarked that most of his stories were not science fiction because they could not happen (except for Fahrenheit 451).  Plausibility of the science in science fiction isn’t a good criteria for a definition, though.  First, it requires that we look outside the work itself to determine if it’s science fiction or not.  Second, it’s easier to look at scientific plausibility as something that affects the quality of the work.  Implausible science may negatively impact the viewer’s willing suspension of disbelief, but it’s not really an assessment of the nature of the work.

You might ask what happens when the imagined science becomes real with the passage of time.  You need look no further than Jules Verne‘s 20,000 Leagues Under the Sea for an example.  What made it science fiction at the time Verne wrote was the imagined submarine, something which is now very real (and when combined with nuclear weapons, far more important to the modern world than we usually realize).  So is it still science fiction?  Yes, it probably is, because the submarine was entirely the product of Verne’s imagination at the time of publication.  We may have to refine our definition a bit, so that it applies at the time of conception.

Now we have the following:  science fiction is fiction which presents science or technology which was imaginary at the time of the conception of the work.  It’s not as simple as it was, and it does require reference to the state of science and technology, which is outside the scope of the members of the genre.  But it’s as good as we’re going to be able to get, I’m afraid.

What if the science is not imaginary, but fully realized?  At that point, we are no longer describing science fiction, but fiction about reality, exactly as it is.  If somone sat down and wrote a story in which a terrorist or revolutionary began attacking shipping on the high seas in a private submarine—the basic plot of Verne’s 20,000 Leagues Under the Sea, of course—the work is a work of fiction, but it is no longer science fiction.  To see what I mean, consider instead a murder mystery, in which capturing the killer depends on forensic science—you know, CSI or Bones or any of the television shows of that type.  We all know in our guts that these aren’t science fiction, and that’s for the simple reason that the science is real.

Is my shiny new definition too broad and too simple?  It’s going to include things which are not normally considered science fiction.

The movie adaptation of Tom Clancy’s novel The Hunt for Red October is a good example (we’re going to stay away from the novel for a minute or two, since it opens up a new set of problems).  In the movie, the Russians equip a typhoon-class submarine with a “caterpillar” or magnetohydrodynamic drive system, and the Russian (okay, Lithuanian) captain of the submarine, the aforementioned Red October, defects to the United States in order to preserve the balance of power.  Now, I’m not saying that no one has ever built a magnetohydrodynamic drive system, because there are a few prototypes out there.  But there isn’t one even close to being able to move something the size of a typhoon-class submarine, let alone move it at reasonable speeds.  It may not be possible at all without some significant improvements in our ability to generate electromagnetic fields or to manufacture really powerful hydrogen fuel cells as a source of power.  And it may be that a drive of this sort, though lacking a propeller and moving parts, would actually be noisier than a standard drive system due to the water flowing through the drive.  Since the movie presents imagined science, it would be science fiction using my definition.  Is that okay?  Personally, I don’t have a problem with it.  It’s always seemed like science fiction to me, so it matches my “gut” feeling.

Now let’s turn to the novel, which is going to give us some problems.  That’s because the caterpillar drive in the novel uses a pump-jet drive system.  Pump-jet submarine propulsion systems are in use today; the American Seawolf class submarine and her little sister the Virginia class both use it precisely because it is quieter than a propeller drive system (the English, French and Russians all introduced their own variants later).  The Seawolf was commissioned in 1989.  Clancy published The Hunt for Red October in 1984.

It sure looks like Clancy used imaginary technology in his novel, which would make the novel science fiction by my proposed definition.  But…what did Clancy know, and when did he know it, about the pump-jet system which was used on the Seawolf, some five years after the publication of the novel?  The pump-jet drive was invented in 1954, and Clancy could have either reasoned out the stealth submarine application on his own or he could have caught wind of a design theory bandied about in military circles—Clancy being something of an aficionado of military technology.  I have no idea which is true.  You’d have to ask Clancy himself, something I am not in a position to do.  The Hunt for Red October (at least the novel) demonstrates one of the places where my definition of science fiction simply fails, because of what we don’t know.  There’s really nothing we can do about that, other than to be aware of the limitation.


I’m satisfied with my definition of science fiction:  fiction which presents imagined science or technology, at least at the time of conception.  You may not be, and that’s okay.  We’re talking about what amounts to a marketing category, after all.

But my definition does have some advantages.  First, it works—everything on the Top Fifty Films List qualifies as science fiction.  Second, it’s fairly easy to apply.  Third, it’s going to be reasonably consistent no matter who applies it.  I warned you at the beginning that these were all things I had in mind when I started my definition.  I don’t think they’re out of line, but they did inform my classification process, as genre theory predicted.

You might argue that the definition is too broad, including things that shouldn’t qualify as science fiction.  Genres have notoriously fuzzy boundaries, and the flavor quality referred to by Isaac Asimov only compounds the issue, but overall, the definition matches my gut instincts.  Lots of people will tell you that Mars Attacks! and Star Wars are not science fiction—but if they aren’t science fiction, then what are they?  That’s the consequence I wanted to avoid, the creation of a new genre called “science fantasy” or something similar (there are critics who do precisely this), which panders to the aspirational definitions of the twenties and thirties and simply ignores popular science fiction.


2 Responses to “What is Science Fiction?”

  1. RE: Tom Clancy, the pump jet was still in Theoretical possibly prototype phase at the time of writing so I think under those grounds you could still class it as sf.

    I was going to be facicious or throw the spanner in the works and suggest that technically science fiction could stand for:
    1) Science that is fictitious.
    2) Fiction that involves science.

    Of course this would incorporate some of the things excluded above such as CSI. Also in it’s broader sense; specifically point 2, this could include the social sciences, life sciences, popular science, hard science… etc so it is quite a minefield out there and I applaud your definition.

    • Thank you for the thoughtful comment—I truly appreciate it. I chose the example of The Hunt for Red October on purpose since it and other technothrillers blur the line between science fiction and fiction, and the difference between the movie and the novel helped me clarify my thoughts, as it turned out!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: