One of the basic working definitions of science fiction as opposed to fantasy hinges on the idea of plausibility: science fiction stories portray events the reader thinks might possibly be able to happen, while fantasy stories portray events the reader knows are impossible. An addendum to such a definition might suggest that science fiction is fiction about science, while fantasy is ... not.

Of course, there are as many exceptions to these rules as there are examples, and the definition of "possible" that underlies the plausibility of science fiction is enormously elastic (as is the definition of "science"). Nonetheless, at least in how a reader approaches a story, there can be something vaguely satisfying about such definitions and contrasts.

But then there are stories about the far future. Is it possible for a story set, say, a million years in the future to be plausible, never mind scientific? Many such stories—Brian Aldiss's Hothouse, for instance—feel more like fantasy than science fiction, and when it comes to categorizing far-future stories, despite proclamations and damnations against the straddling of taxonomic fences (the best I know of being James Blish's argument against "science-fantasy" in More Issues at Hand), what we're usually dealing with are stories that utilize the props of science fiction in utterly implausible and unscientific ways.

Why does it matter? Normally, I would say it doesn't, but what we have with Gardner Dozois's new anthology One Million A.D. is one example after another of why people writing about the far future would probably be better off thinking of themselves as fantasy writers than as science fiction writers—the attempt to write plausible science fiction has, in most of these stories, limited the authors' imaginations and forced their tales into dullness. One Million A.D. includes six novellas by Robert Reed, Robert Silverberg, Nancy Kress, Alastair Reynolds, Charles Stross, and Greg Egan—experienced and talented writers, each—but despite the opportunities offered by the book's theme, only half of the stories are the least bit memorable.

Some of the most imaginative and adventurous science fiction stories of the last fifty years have been tales of the far future. Think of Cordwainer Smith's bizarre narratives of a future mythos, or Frederik Pohl's "Day Million," with its magnificently odd opening: "On this day I want to tell you about, which will be about a thousand years from now, there were a boy, a girl, and a love story." Not one sentence in One Million A.D. contains the same power as Pohl's to plunge the reader immediately into a strange and unfamiliar world. Along with its inventiveness, what distinguishes Pohl's story is the strength of its voice—the narrative voice in "Day Million" is distinctive for its colloquial tone (and casual sexism) as well as its occasionally nonstandard syntax and diction. In contrast, all of the six novellas in One Million A.D. are told in the same flat, interchangeable prose. There are slight variations of tone and vocabulary, but in general the sentences from one of these stories would sound just the same in any of the others.

A case can be made for flat prose being useful for creating a sense of plausibility when what the prose describes is bizarre and unlikely. In at least one of the stories here (Robert Reed's), I think that's probably exactly what's going on. But a writer who chooses not to use language itself as a tool for adding depth to the imagined world must then strive toward truly creative settings, characters, images, and events if the story is to be anything other than ordinary, because the danger of flat prose is that it can drag everything it describes down to its own low level.

Aside from the prose, there are other similarities between most of the stories—all of the plots are structured as either mysteries or quests, half of the stories imagine futures in which humans can remake their bodies (and sometimes other elements of the material world), and most of the stories spread themselves thin in an attempt to render vast views of the universe.

Gardner Dozois's short introduction to One Million A.D. raises expectations the tales themselves can't fulfill. After a general overview of the history of far future SF, Dozois says,

Here in the early days of the twenty-first century, I decided that it was time to get a twenty-first century take on what the far future was going to be like, and so I challenged a group of daring visionaries to imagine life in One Million A.D.: a time so far ahead that the human race and the Earth itself will have been altered almost out of recognition, and all our history and culture, everything we are, everything we know and cherish, will have faded to dim and half-forgotten mythology—if it's remembered at all. (p. x)

This is a recipe for exciting fiction, but it is a recipe few of the stories in the book live up to. The view from the twenty-first century turns out to be pretty much the same as it was at the end of the twentieth century, with the futures imagined here either being post-apocalyptic or post-human, seldom in any way that hasn't at least been hinted at many times over the last few decades. "A group of daring visionaries" is a particularly unfortunate label, because a kid doing bong hits in his basement is more visionary, and certainly more daring, than most of what's available here.

Take, for instance, Greg Egan's "Riding the Crocodile," a story that begins well, with a light tone and immense scope: the tale of two immortal creatures who have, over their 10,309 years of marriage, done everything they want to do, and have decided to let themselves die, but not before attempting "something grand and audacious." Namely, getting in touch with a section of the galaxy, a civilization known as the Aloof, that seems to have deliberately kept itself out of contact with the rest of the known universe. The story is structured mostly as a mystery—why are the Aloof so aloof, and how can they be contacted? The problem is that there is little reason for a reader to feel any stake in this mystery. We don't care if the two immortals are in peril, because their long lives have made them bland personalities, and there's no reason anyone should care whether they live or die. We don't care who the Aloof are, because the world of the non-Aloof is so plentiful and vast as to be inconceivable. There's nothing much else to focus on in the story; in one-fifth the length, Egan could have accomplished as much. "Riding the Crocodile" isn't a bad story, it's just not a very good one.

It's pointless to beat up on mediocre stories, though, so let's talk about the ones that are the least yawn-inducing, of which there are three. Alastair Reynolds's "Thousandth Night" is the weakest of the three, because Reynolds demonstrates little ability to create characters who are personalities instead of cliches, but the story of near-immortals remaking the universe has some fun set-pieces, and the central mystery at least keeps the story from being tedious. Charles Stross's "Missile Gap" is a better tale, because the central concept is crazier than any other in the book, and the ending delivers some amusing folderol with time warps. Stross creates a kind of alternate past-within-the-future where an Earth with a Cold War that got slightly hotter than ours did is peeled away by aliens and put on a gigantic disk floating through space. The weakness of the story is that of many of the stories here—most of the events are little more than an excuse to give us a tour of the landscape—but at least in this case the landscape provides some good sightseeing.

Robert Reed's "Good Mountain" is the most satisfying story in the book, the least awkward and most vivid. Reed presents a world of free-floating islands in a sea of poisonous water, a world of entropy and decay and metamorphosis where giant worms with hollowed-out chambers provide transportation for, among others, travelers seeking to escape realms destroyed by deadly accumulations of methane. The story works for a number of reasons, not least of which is the dark ambiguity of the ending—too many of these stories end with neat, tinkly finishes, but Reed is not afraid of the shadows within the world he has created, and he does not force resolution where no resolution is possible. The writing is also a strength; every other story in the anthology suffers from superficial or ineffective descriptions, but much of Reed's descriptive writing, particularly in the first sections of "Good Mountain," carefully constructs the world of the story through evocative imagery that is specific and yet expansive:

Unable to grow, the island shrank. Hungry, it drank dry its sap reservoirs. It could have brushed up against the Continent again, perhaps several more times, but some current or chance storm always pushed it away again. Then it wandered, lost on the dark face of the world. The evidence remained today inside its body. Its oldest wood was full of scars and purple-black knots—a catalog of relentless abuse brought on by miserly times. Not even a flicker of sunlight fell on its bleached surface. Starving, the island digested its deep-water roots and every vein of starch. Saprophytes thrived on its surface and giant worms gnawed their way through its depth. But each of those enemies was a blessing, too. The tallest branches of the saprophytes caught the occasional breeze, helping the increasingly frail island drift across quiet water. And the worms ate so much of the island that it floated easily, buoyed up by the air-filled caverns. (p. 8)

Such a passage could easily have been a clunky paragraph of exposition (all of the other stories in the book suffer from just such clunky paragraphs), but Reed turns what might otherwise have been a weakness into a strength by keeping the sentences focused on sharp, specific details that build a wide view of the imagined world in the reader's mind. Each bit of information adds to the immediate image while also adding to our knowledge of the story's setting, and each sentence contains action—the description is vivid because everything being described is in motion.

Though "Good Mountain" is not reliant on specifically twenty-first-century knowledge, nor is it a particularly daring story, it does live up to the rest of Dozois's description of the stories in the book, certainly more so than any of the others, and passages such as the one above do have visionary qualities. At the end of his introduction, Dozois indulges in utter hyperbole, saying the stories he's collected are "some of the most vivid, evocative, entertaining, and mind-stretchingly imaginative science fiction you're ever likely to enjoy." Not even Reed's story lives up to such a ridiculous description, but at least it provides some of the imaginative pleasures we might expect from a story set in an impossibly far-off future.

Matthew Cheney's work has appeared in a wide variety of venues, including English Journal, Locus, SF Site, Rain Taxi, Failbetter.com, and Rabid Transit. He writes regularly about SF and literature at his weblog, The Mumpsimus, which was nominated for a 2005 World Fantasy Award. You can also find his work in our archives.