Don't tell anybody, but science fiction no longer exists.
Let me rephrase that, because though the statement is true, it's not entirely accurate. Try this: written science fiction has become two things—a genre and a style; the audience for the former is small and shrinking, while the audience for the latter is large, growing, and doesn't know of its own existence.
I tend to use a fairly conservative definition of the word genre, a word that can mean an awful lot of different things to different people. For my purposes, genre signifies a specific type of writing, one that may have some grey areas, but has enough boundaries for readers most of the time to be able to say whether something is or is not a part of the genre.
A good and usable definition of science fiction comes from Stanley Schmidt, editor of Analog, who said in a 1999 interview:
[S]cience fiction is simply fiction in which some element of speculation plays such an essential and integral role that it can't be removed without making the story collapse, and in which the author has made a reasonable effort to make the speculative element as plausible as possible.
(The grey area comes when you try to define "speculative" and "plausible" and then attempt to convince anyone to agree with you.)
A style is different; a style is simply a way of saying something.
Certainly during the second half of the twentieth century, science fiction at its best was sometimes more style than genre. Even in the middle of the century, it could accommodate Hal Clement and Philip K. Dick, Jack Vance and Kate Wilhelm—vastly different writers, some of whom wrote work that wouldn't fit very comfortably within Stanley Schmidt's definition of SF, but all of whom were clearly a part of the SF community, even if the fans of one had trouble understanding the enthusiasm displayed by the fans of another. The 1960s and 1970s further stretched SF from being a genre toward being a style, with Gregory Benford and Larry Niven getting nominated for awards alongside Samuel Delany, Ursula Le Guin, and R.A. Lafferty. Judith Merrill, Michael Moorcock, and Harlan Ellison edited books and magazines that seemed to acknowledge no genre boundaries at all, and the work of J.G. Ballard, Thomas Disch, M. John Harrison, Norman Spinrad, and many others expanded what could be written with SF tropes, using spaceships and aliens and all the technological claptrap of everyday life to speculate more about the present than the future. All the while they did this, though, they demonstrated a clear understanding of science fiction as a continuing discussion, as a body of literature with a past, and with certain conventions of reading and writing—thus, as a genre.
Certainly much changed in the 1980s and 1990s, but the biggest change was not the advent of the cyberpunks or a new group of writers like Greg Egan and Stephen Baxter who tried to put the science back in science fiction. No, the real story was that science fiction as a style began to be exploited by many writers who never got labelled as SF writers and who had at best an indifference to its history—while at the same time, science fiction as a genre pulled back from the earlier experiments and settled into recycling the old techniques, subject matter, and definitions.
In the February 2005 issue of Locus, Jonathan Strahan looks back on 2004 and says:
The urgent need for SF to explore new directions was evident in Robert Silverberg's Between Worlds, the best SF anthology of the year, which contained six strong novellas, most of which could have been written 30 years ago.
In the same issue, Gardner Dozois says of Silverberg's book:
As an old dinosaur, solidly retro and uncool in his tastes, it struck me as a nice change to see an anthology that featured nothing but solid center-core SF, instead of the trendy genre-bending and mixing that many of the year's other anthologies attempted to one extent or another.
Between Worlds, then, is perfectly good as a genre anthology, doing what has been done for decades now, and doing it well. The same is true for the major SF magazines—they have relatively clear lines for what sorts of things are and are not SF to them, and their readers expect the editors to deliver stories that fit within those lines. According to the same issue of Locus from which I quoted above, in 2004 Analog's circulation dropped 18%, Asimov's dropped almost 9%, and Fantasy & Science Fiction's circulation dropped 11.8%.
Meanwhile, SF as a style, or at least a tendency, pops up in bestselling thrillers and the most literary of literary novels published by university presses, while novels like David Mitchell's Cloud Atlas get nominated for both the Nebula and the Booker Prize.
Consider this: in the 1960s and 1970s, one of the most innovative writers in America, David R. Bunch, was published in magazines like Fantastic and Worlds of If and Fantasy & Science Fiction, his bizarre and challenging little stories surrounded by the stories of people such as Robert Silverberg and Poul Anderson.
The ripples of the New Wave flowed into the mainstream, but not the mainstream of SF. The New Wave writers, it turns out, were ahead of their time in mixing the subject matter of science fiction with the techniques of modernist writers, fusing scientism with surrealism, escapism with absurdism, William S. and Edgar Rice Burroughs. Today the lessons learned from the experiments of New Worlds and Dangerous Visions are most clearly visible not in the pages of Asimov's or F&SF, but in the fiction published by The New Yorker and Harper's, because today's young turks are influenced by both high modernism and the super-technologized world we live in. The most celebrated contemporary writers enlist both Nietzsche and Nintendo in their quest for the real.
Maybe "style" isn't the word I want. Maybe "pose" is more accurate. But "pose" suggests artificiality and shallowness, and I don't think the writers who use science fictional terminology, settings, and set pieces are being either artificial or shallow—they are, in fact, writing some of the most vital and imaginative contemporary literature. Literature that addresses both what it feels like to be alive today and what it might mean to be alive tomorrow. This is a type of fiction we need, and though it once resided mainly in the SF ghetto, its ascent to the highest realms of the literary world should be celebrated.
Nonetheless, I feel sad for the ghetto, because the expanded imaginings of mainstream writers have paralleled a narrowing of imaginations from the major SF publishers.
Science fiction could have remained a flexible genre, and certainly there are instances now when it does, but on the whole the lovers of SF as a genre have become so distressed by the current scene that they have resorted to tactics that are likely only to further marginalize them. Instead of encouraging writers who have a sense of the history and substance of genre SF to experiment with form, language, and even the basic meaning of fiction, today's writers are given the message (most loudly through rejection slips) that to write science fiction means to write as if nothing but the gadgets had changed since John Campbell's heyday at Astounding in the 1940s. Consequently, the very writers who could revitalize SF and make it a less moribund genre go off and do other things and find audiences that actually appreciate their creativity.
Today, David Bunch would be rejected by the major SF magazines and published by literary magazines such as Conjunctions and Fence. He would be compared to writers like Thomas Pynchon, David Foster Wallace, and George Saunders, and he would probably win a MacArthur "Genius" Grant. Young, celebrated writers like Matthew Derby and Ben Marcus would acknowledge the debt they owed him. "A science fiction writer?" people would say skeptically when anyone suggested that that was a label Bunch deserved. "No, he just writes about the implications of technology on what it means to be human, creating postmodern fables of alienated identity. That's not science fiction."
No, it's not. Not anymore.