Yes, but Is It Good?
It’s a question we skirt around all the time as science fiction critics.
We’re comfortable asking: can it be bad science fiction but a good novel? That one’s easy: of course it’s possible. There’s any number of mainstream attempts at science fiction out there that have all the qualities we associate with well-wrought fiction: finely delineated characters, lyrical prose, and so on and so forth. But when it comes to the science fictional idea the whole thing collapses in a mess of idiocy, of over-used gimmicks and under-developed thought.
Turn the question round, however, and we start to stumble. Can a work be good science fiction but a bad novel?
When you consider the question in the abstract it’s simple. No, of course it’s not possible. We want it all: great ideas and great literary values. After all, how are you going to get those rich, complex, fascinating ideas across to your reader if you can’t drive the language properly? There’s a certain arrogance in this position, of course. Science fiction has to be better than any other form of literature because it does all the stuff the mainstream does, with extra added ideas.
However, start to name individual writers or novels and this position starts to come unstuck. What about Isaac Asimov? The only thing female about Susan Calvin in all the robot stories is her name, yet those stories shaped the genre despite the feeble characterisation. How about Robert Heinlein’s Starship Troopers? It has all the literary sensibility of a Tonka toy, but it is an important work of science fiction nonetheless.
Or let’s take a current example: Incandescence by Greg Egan. I don’t know if Egan could ever make it as a literary novelist, but he’s the ideas man par excellence of contemporary science fiction. A large part of this new novel involves complex ideas from mathematics and physics being rediscovered and explained in simple, non-mathematical language that has even someone like me, who has difficulty with simple sums, feeling like I can grasp what is going on. It is the very model of a modern hard sf novel.
Take the "hard sf" out of that sentence, however, and you’re on dodgy ground. Plotting is rudimentary at best, particularly the feeble trigger that sends Rakesh and Parantham off on their space operatic quest into the heart of the galaxy. As for character, Roi and Zak and the other insectoid creatures on whom Egan focuses most of his attention are hardly brilliantly defined individuals. Though that is largely the point, given the nature of their race and civilisation.
So we have a book that thrills as science fiction, but barely satisfies as fiction. Do we count that success? Or failure?
Actually, what we should do is acknowledge that we are asking the wrong questions. When we ask if it is possible for a work to be good science fiction and a bad novel (or even the other way round) we are making an assumption that there is some absolute standard for what makes good science fiction, and some absolute standard for what makes a good novel. There is neither.
I am tempted to describe a novel (or indeed any piece of writing, fiction or nonfiction) as no more than a vehicle designed to convey a cargo (story, ideas, impressions) from point A (the mind of the writer) to point B (the mind of the reader). Such a description, of course, raises more than enough questions of its own to have any self-respecting academic foaming at the mouth, but I’m not inclined to go into here; let it stand for the moment at least. None of us can ever know both point A and point B (in a very real sense, no one can be the reader of their own writing), and most of us only ever know point B, so our assessment of the efficacy of the conveyance (or the communication, if you will) must always be partial. That’s part of the territory, we must live with it.
Good writing, therefore, can perhaps best be described not in terms of verisimilitude or poetic imagery or any of the other normal attributes we associate with the term. Such attributes, rather, delineate distinct forms of good writing, particular genres if you will. No, good writing, in broad and general terms, should be described simply as whatever writing best conveys the novel’s cargo. So, for certain ideas you may need a very simple, unadorned, fact-laden prose; certain images might be better conveyed by a prose that aspires to be poetry. Similarly, leaden and repetitive writing is bad if it is put at the service of a story that is meant to carry the reader away with wonder; while prose that is heavy on the adjectives and allusions doesn’t really work if you are trying to get across something crisp and clear and unequivocal.
This is a rather dull, prosaic, empirical approach to good writing: it’s what works best for the story being told. Goodbye to absolute standards of quality, and goodbye also to any notion that "good writing," however defined, is in and of itself enough to mark any work as good. Okay, there may be some books whose sole intent is to make the reader swoon at the beauty of the words, but they are few and far between. Most works of fiction (and, indeed, most works of nonfiction, most dramas, most poems, most business reports, most blog posts, what have you) have something more in mind than the enchantment of language alone. What that something more might be will, of course, vary. It might be story or character or description or sense of place or facts or argument or a myriad other things, and the way the prose is used to convey that will inevitably vary also. And that does not mean that one type of writing must necessarily be used for one type of purpose -- far from it. The unutterably beautiful and the deliberately flat might each work perfectly well to tell a story or present a set of facts. On the contrary, what I am trying to stress is that there are any number of forms that good writing might take, and they depend on any number of other things that are going on in the work.
So let us go back to those original questions: is it possible for something to be bad science fiction and a good novel, or to be good science fiction and a bad novel? The questions are meaningless. They only work if you assume that there is some absolute standard of "good novel" that holds whatever else that novel might be doing, and that is just ridiculous. You can ask: is it a good novel? Or you can ask: is it a good science fiction novel? (In reality, the two amount to the same thing.) But you cannot set one against the other. If the elements that make the story science fiction fail, then it is a bad novel regardless of how good the writing or the characterisation or any other things that go around it might be, because you read a book as a totality not as a set of distinct and separable parts. Similarly, if a science fiction novel is full of great ideas but is written in a prose style that either makes interesting ideas dull or that so overwhelms the ideas with extravagant prose that the ideas are lost, then it is a bad novel.
Science fiction sets high standards for itself, because we value the ideative content of the work as much as everything else that normally goes to make a good novel. But that only means it is easier to fail, since there are more things that can go wrong. Yet those high standards are not different standards, all that is required to make a good science fiction novel is exactly what is required to make a good novel of any other genre: the prose may be plain or ornate, but it must make what is important in making the novel work sing in the mind of the reader.