Against a Definition of Science Fiction
How we define science fiction dictates how we determine where the genre began and of what it consists. But is the genre really susceptible to definition?
When I called my collection of essays and reviews What It Is We Do When We Read Science Fiction, I was struggling toward something I could not fully articulate. I don't know what is involved in reading science fiction, because I don't know what science fiction is.
There was a time, not so long ago, when I was quite clear what science fiction is. I could pick up a book from the shelf and know, with no real doubt or confusion, that it was science fiction. That certainty is with me no more, not because science fiction has changed (it has, of course, in many and complex ways, but not in a way that is directly relevant to this discussion), but because my relationship with science fiction has changed. And the nature of that change lies in my increasing criticism of and theorizing about the genre.
When the earliest of the pieces included in my book was written, I was simply reviewing sporadically for a number of publications. I had no notion that my reviews might ever be included in a book; I had even less notion that I was in some way expressing a coherent critical response to the genre. I was simply a reviewer; all that was coherent was that the views expressed came from one person, one set of experiences, and since those experiences changed with each new science-fiction novel I read, it was an evolving rather than a coherent view I imagined myself expressing.
By the time the possibility of the book first emerged, my views about the nature of science fiction had changed many times. I had begun reading critical theory, taxonomies of science fiction, histories of the genre. The more I read, the less certain I became about what it was I was reading. And yet the more certain I found other critics and commentators.
I am not going to repeat the long and tedious story of getting the book into print, but I will note that not only was there a gap of something like eight years between selling the idea of the book to one publisher and the finished thing actually appearing from a different publisher, but there is also an awfully long time represented within the book. I think there's a gap of something over twenty years between the earliest piece in the book and the most recent.
Given the time scales involved, it would be ludicrous of me to try and claim there was one guiding principle behind the book. There wasn't; there couldn't be. I change my mind about science fiction on a regular basis. Every time I read a book that seems to settle one view of the genre, then the next book I read is guaranteed to undermine it. Over the years I have come to realize that this is one of the things I like about science fiction - the fact that it confounds expectations, makes you think again.
But more and more, as I rather belatedly started to write essays with the expectation that I might one day try to put them into a book, and more especially after I started cutting out some of the pieces I'd originally gathered and replaced them with some newer, hopefully better pieces, I realized there was one serious and interesting topic that I did keep circling around. It's a topic that is almost but not quite captured in the title: What is this thing called science fiction?
Actually, the title of the original essay, and hence of the collection, was inspired by Raymond Carver's What We Talk About When We Talk About Love. I sometimes think I should have been truer to the inspiration and called my book What We Talk About When We Talk About Science Fiction.
The slipperiness, the confounding, undermining character of science fiction may be one of the joys of the genre, but it makes it bloody difficult when you're trying to write about it.
I have to be careful here. A little while ago someone wrote in his blog that all of British genre criticism seems to be obsessed with taxonomy. Well there's a certain amount of truth in that, but there's also a great deal of untruth. An interest in taxonomy is by no means confined to British genre criticism, nor is it the sole direction of British criticism. Nevertheless, at least two books appeared in 2008 from British critics, mine and Farah Mendlesohn's Rhetorics of Fantasy, that seem taxonomical. Though I suspect that neither of us intended them to be seen that way, and I certainly thought of mine as more anti-taxonomical.
So, without actually wanting to produce a taxonomy of science fiction, I did find myself returning again and again to the question of what I was actually writing about.
You know, there are an awful lot of supposed definitions of science fiction out there. They all contradict each other, and for any definition of science fiction I suspect that each and every one of us could come up with at least one novel or story we all agree is science fiction but that doesn't fit the definition.
Darko Suvin's characterization of science fiction, the definition that is almost universally taken for granted by academics working in science fiction, actually allows for a book to be science fiction at one point and not science fiction at another, later point because of changing views of science. To be honest, that doesn't strike me as being very useful.
The core of Suvin's characterization is his notion of “cognitive estrangement.” For a long time I was happy to spout “cognitive estrangement” as the best definition of science fiction; I mean it sounds really clever, doesn't it, even if you're not exactly sure what it means? And when you do try to disentangle what it's saying, it does seem to apply to science fiction.
My problem came as I tried to juggle with all the different things that science fiction did, all the different ways we approach the genre, and I tried to find some unifying theory that tied it all together. It started to bother me that Suvin's “cognitive estrangement” applies so broadly to science fiction only because it applies to an awful lot of other things as well. It's all very well to say that “cognitive estrangement” describes what science fiction does; that's fine, and it's probably, mostly, true. But to take the next step and say that it defines science fiction can only be true if it accurately describes everything that we recognize as science fiction, but equally accurately excludes everything that we recognize as not science fiction. Unfortunately, cognitive estrangement doesn't do that.
Let's take a step back. “Estrangement,” as Suvin tells us, is a term taken from a bunch of Russian philologists and critics in the early and middle years of the twentieth century who are collectively known as the Russian Formalists. What it means when they use it is the way poets employ words to slow us down. As they see it, a poem isn't meant to be smooth, so that you glide across it quickly so you don't really notice it. A poem is meant to be jagged, to have words that we don't immediately recognize, or that seem to come from a different vocabulary (in T. S. Eliot's The Waste Land, for instance, the way the generally high-flown language is suddenly interrupted by the Cockney women in the pub), or that throw up unexpected images (again in Eliot, London as a patient etherized upon a table). These jagged words estrange us from the poem, forcing us to read more slowly, to look more closely at the rhythm of the words or the ideas being presented.
Cognitive estrangement, therefore, is a jagged idea or image that makes you stop and look again at what you know, that makes you look more closely at the world being presented: “The door dilated,” for example, in Robert Heinlein's The Door into Summer. Or take Samuel Delany's notion of the concretized metaphor: “She turned on her side” could mean she rolled over in her sleep, or she threw a switch. That's cognitive estrangement.
Yeah, that's science fiction, isn't it?
Except, think what happens when Jane Austen begins a novel “It is a truth, universally acknowledged . . .” If it was universally acknowledged, she wouldn't have needed to say it. She probably wouldn't even have thought to say it. She is making her readers stop and think again about the world that is being presented. “Hey, I don't acknowledge that. What is she talking about?” It is cognitive estrangement. Satire is cognitive estrangement. Practically every joke is cognitive estrangement. When Monty Python has crucified martyrs singing “Always look on the bright side of life,” it is cognitive estrangement. When a science-fiction critic sits here and tells you that Jane Austen uses cognitive estrangement, it is cognitive estrangement.
So cognitive estrangement is useful, but not really as useful as all that when you're trying to tie down what is different about science fiction. And all the time I'm trying to nail this down (for some reason the image of nailing jelly to the wall comes to mind), I keep coming back to Damon Knight's ostensive definition: science fiction is what I point to when I say science fiction.
You see, the ridiculous thing is that science fiction is so easy to recognize. We all know whether a book is science fiction or not when we read it, don't we? Oh there's always some other idiot who'll say, no, it's really fantasy, or no, it's really horror. But ignore that idiot: in our own minds we're generally pretty sure. So why is it so difficult to define?
The answer I eventually came up with - and believe me, this is the sort of thing you can skirt around for years without recognizing, then it sort of hits you in the face and you go: duh, of course; in fact, when it did occur to me I thought it was so blatantly obvious that I nearly didn't bother writing it down - was that science fiction isn't one thing. It's lots and lots of different things that we gather under the umbrella of science fiction. Science fiction is a collective term. But none of these different things have exactly the same characteristics, which is why there can be no one set of characteristics that define a science-fiction story.
And once I had that idea, I could go back to Damon Knight's ostensive definition and combine it with Wittgenstein's notion of family resemblances (which is one of the few things I remember vividly from my philosophy degree years and years ago) and come up with a way of characterizing SF that seems to me both accurate and useful.
Okay, for those of you who have got this far through life without ever encountering Ludwig Wittgenstein's Philosophical Investigations: family resemblances comes up when he's talking about how we use the word “sport.” We know what a sport is: it involves two teams, except where it doesn't; and they use a bat or racket or club or none of the above to hit a ball or shuttlecock or puck or none of the above; it demands strength except when it's, say, diving, which demands grace; or just possibly like shooting, which demands accuracy; it demands speed except in a sport like snooker or darts; its participants are invariably at the peak of physical fitness unless they are golfers who are often overweight, overage, and smoking. You get the picture. But we recognize what things are sports because they have family resemblances with other sports. Similarly, we know when we are reading a work of science fiction because it has family resemblances with other science fictions.
There's a bit more to it than that, but essentially that's got me as far as the essay I called “On the Origins of Genre,” which is at the heart of my book. Actually, it comes pretty close to the beginning of the book even though it was one of the last things written. So when you're reading it, perhaps I could suggest that you consider the later chapters not as a consequence of what went before but as part of what fed into it, as background.
And in a sense that should be the end of my story, except, of course, that it's not. I have presented a relativistic view of science fiction that I am comfortable with, but it is not an unproblematic view. And I have barely begun to work my way through the issues raised.
If we accept this view of science fiction, for instance, then a lot of our understanding of genre comes down to the way we read. The ways we interpret a text, how we recognize the clues within the text, what prejudices we bring to the reading, what experience of other genre works we bring, can all have an effect on whether we identify a book as belonging to one genre or another. This is an area that has been fascinating me lately.
When I taught at the Foundation Masterclass, I asked the students to read four works: two novels, The Translator by John Crowley and Light by M. John Harrison, and two stories, “Magic for Beginners” by Kelly Link and “The Disappearance of Elaine Coleman” by Steven Millhouser. The reason I picked these four is that they are all problematic in terms of genre identification.
Naturally, I didn't say why I'd picked them ahead of the game. When I asked the students why they thought I'd picked them, they concluded it was to torment them. Curiously, none of the four was much liked before we began the discussion, even though a secondary reason for choosing them was that they are all personal favorites of mine. Still, as a critic you get used to no one else sharing your opinion.
But the result of the discussion in the masterclass was fascinating. One of the students was an American who had quite a bit of knowledge of the political history of the early 1960s, and he read The Translator as being a straightforward historical novel with no fantastic elements whatsoever. Others had assumed it was fantasy, simply because it was by Crowley.
One of the students really hated Light as a work of science fiction. So I took us through a variant reading as a work of psychological realism in which the future episodes are representations of the psychosis of the protagonist, and suddenly he loved the book.
The Millhouser became everything from a trite crime mystery to a profound meditation on collective loneliness.
But it was the discussion of “Magic for Beginners” that was the real revelation. When we started talking about it, there wasn't one of us who admitted to understanding the story. By the time we'd finished we had looked at the way Link uses frames within frames, we had considered the identity of the occasional “I” narrator, we had asked why the story starts where it did (since none of the readers paid any attention to the very first paragraph), and we had worked backward from the ending to try and find out why it finished at that precise point. The story was still mysterious, but we had all changed the way we looked at it.
In fact, by the end of the session every one of the students said they had changed their opinion of at least one of the works we discussed, and all said they liked the stories better.
One thing the session did, for me, was confirm my sense that there is no such thing as pure genre. Everything is capable of being read in different ways. So the way we read a work, sometimes the way we choose to read a work, is crucial in determining how we identify it.
Now this is something I'm only beginning to work my way through. I have reached no conclusions, I am not altogether sure where it leads me, and I am far from being ready to commit anything to writing. But it is something I am finding increasingly interesting. For instance, what do our favorites say about how we see genre? By favorites, I mean those works we read over and over again, or those authors whose new book we pick up the moment it is published. If we read a book persistently, as some people read John Crowley's Little, Big or J.R.R. Tolkien's The Lord of the Rings religiously every year, we are presumably not exhausting the book. Or are we? Is it the comfort of sameness, or the challenge of constantly finding newness even in the familiar? Whichever, this repetition is presumably doing something to shape our picture of how the genre should be. Is someone who reads Little, Big because they want Edgewood to be the same place every time reading the same genre as someone who reads The Lord of the Rings expecting to find a world so big that there is always something new to discover? I can envisage a genre that encompasses both works, but can you envisage a genre that encompasses both ways of reading?
Or again, I will snap up a new book by John Crowley or Steve Erickson as soon as I can get my hands on it, and read it at the very first opportunity. Now that, surely, is going to shape the way I understand the fantastic. But in what way?
I don't know the answers to any of these questions. At the moment I am barely clear what questions to ask. But I find them fascinating, and one thing is for sure, I wouldn't even be considering these questions if I hadn't embarked on the wrestling match with the genre that grew into What It Is We Do When We Read Science Fiction.
Editorial note: This is a revised version of a talk originally given to a joint meeting of the Science Fiction Foundation and the British Science Fiction Association.