September 2006

The Importance of Maps
by Sarah Monette
www.sarahmonette.com

Having completed her Ph.D. in English literature, Sarah Monette now lives and writes in a 99-year-old house in the Upper Midwest. Her first two novels, Melusine (2005) and The Virtu (2006), have been published by Ace Books, with two more novels in the series to follow: The Mirador (2007) and Summerdown (2008). Her short fiction has appeared in many places, including Strange Horizons, Alchemy, and Lady Churchill's Rosebud Wristlet, and has received four Honorable Mentions from The Year's Best Fantasy & Horror.

When I was twelve, I wrote a fantasy novel. It's exactly as bad as you would expect, but what interests me, lo these twenty years later, is how aware I was of the conventions and apparatuses of the genre I had chosen. I knew fantasy stories are quests; I knew they have to have a MacGuffin; I knew the heroes have to go up against the Ultimate Evil. And I knew that if you invent a secondary world, you have to draw a map.

I still have the map: an amoeba-ish blob labeled (in my best and fanciest twelve-year-old handwriting) MEGAR and divided into countries, each with its capital city carefully marked: North and South Carturia, Re Jea, Rombala, Ankar, and the Evil Land with its Dark Castle. It's a rudimentary sort of world-building, but it's world-building all the same, a recognition that writing secondary world fantasy involves the author in a project that is considerably more complicated than just telling a story.

1. WHAT IS IT AND WHY DO IT?

World-building is the art of making a reader believe in a world of which they have no experience. Evoking the world your characters live in is not world-building; it's grounding (itself an art). World-building comes in when the world you're evoking is one you believe to be alien to your intended audience.

This is really easy to figure out in science fiction and fantasy, because we're writing in worlds that don't exist outside our own fevered imaginations; it's easy in historical fiction taking place more than, say, fifty years ago. It's tricky in fiction set in a time contemporaneous with the author, or in the recent past, and depends entirely on how the author approaches the temporal/cultural setting she has chosen.

But we're talking about fantasy and science fiction, in which world-building is a given of writing a story. In fact, one way to define the broad genre of speculative fiction is that, in it, the setting and the plot are mutually dependent. Setting in speculative fiction isn't merely setting; it's world. And world is crucial.

Why should this be so?

All fiction is the art of making us believe, temporarily and provisionally, in that which is not real. Realistic fiction, which takes place in a recognizable facsimile of the real world, and in which nothing happens that could not happen in the real world, can bolster its fiction with the accumulation of a thousand thousand tiny details of everyday life. The writer doesn't have to invent them — and, in fact, shouldn't — but observe them and describe them. Don't get me wrong. I'm not saying that's easy. In fact, I would say one of the major reasons I write fantasy and science fiction is that I'm not observant. It's much easier for me to make things up than simply to notice them.

But the fact remains that the plausibility of realistic fiction is built on the empirical world that the author shares with the reader. Even a bad or vague description of driving south on I-95, or staring at the refrigerator cases in the frozen foods aisle at the local grocery store, is still a description that the broad cross-section of American readers can relate to.

On the other hand, the speculative fiction writer deliberately puts herself in a situation where she has to describe what it's like to fix a broken gravity generator on an airless asteroid or to barter with wood elves for passage across a river. The details of those scenes must come from the imagination. What do elves smell like? What pirate radio station do you listen to inside your pressure suit to keep the big black emptiness of space from squashing your psyche like a bug?

Now, it's perfectly possible to write speculative fiction in which the background is nothing more than canvas and plywood: stage sets. It's even possible to write quite successful speculative fiction that way. But I would argue that it is not possible to write good speculative fiction, because the essence of speculative fiction is in imagining worlds that have never been.

2. DISCRETION IS THE BETTER PART OF RESEARCH

It is true, however, that the speculative fiction writer need not make his world out of merely airy nothings. He has the whole wealth of human history to consult for models for the civilizations he creates. And he has the power to pick and choose, to make his civilization a melange of the ancient Aztec and the ancient Greek, for instance, or to play alternate-history games such as: what if North America was discovered by Asian rather than European explorers? It's speculative fiction; if you can make it convincing, you can do anything you want.

But there are still pitfalls. (There are always pitfalls, and tiger traps with angry tigers in them.) Elizabeth Bear tells a story about — when she was researching to write an Elizabethan historical fantasy — reading a variety of other Elizabethan historical novels, both fantasy and realistic, and coming to recognize that the authors had all read the same three nonfiction books. This problem is less glaring for the author creating a secondary world, but it is still worth keeping in mind. Chase footnotes; try to find books off the beaten path. If nothing else, you may find that they have the most interesting and idea-generating tidbits in them. Don't be afraid of trivia. In world-building, tiny details can mean a lot. Also, when you're researching, take good notes. Make as few assumptions as possible about what you will and won't use, and do not rely on your memory to cough up the fact you need when you need it.

But if lazy research is Scylla, Charybdis is what the Turkey City Lexicon calls "I've Suffered For My Art." Just because you know something doesn't mean your reader needs to know it, too. Your world-building will in fact be more convincing if you don't stop to explain every reference or peculiar custom, because that's the way the real world works. There are all kinds of things about even our own culture(s) that we don't understand or can't explain or don't know the origin of. Give your invented society that same leeway.

And perhaps most importantly, do your own work. If you're going to do research, it had better consist of more than reading other people's novels. Don't borrow world-building from other authors, and especially, please, don't borrow wholesale from Tolkien.

3. TOLKIEN AND HIS DISCONTENTS

The great master of secondary-world fantasy, and thus of world-building, is J. R. R. Tolkien. But it is important to recognize the ways in which Tolkien was a special case.

It is famously said of Tolkien that he invented Elves in order that his languages would have someone to speak them. While tidy, as most epigrams are, it is a distortion, but it points toward the truth, which is that Tolkien was far more interested in his world-building than he was in writing salable novels. Creating Middle-Earth was his lifelong project; The Hobbit and The Lord of the Rings are in the nature of fortunate by-products.

Most writers today find themselves in the opposite situation. What we want to do is write stories; the world-building may be an integral part of that process, but it does not govern what we do.

This is one reason slavish imitation of Tolkien is a bad idea. There are others.

When Tolkien wrote The Lord of the Rings, there was no tradition of secondary-world fantasy. Some authors made up worlds (H. P. Lovecraft, Robert E. Howard, Lord Dunsany, E. R. Eddison), but each of them was very distinctly doing his (or her — Hope Mirlees, C. L. Moore) own thing. The fantasies they wrote were personal, each one built from scratch. (And of course they corresponded and read each other's work and were influenced by each other, but still — each writer dancing to a different drumbeat.) What Tolkien did, without at all meaning to, was to create a paradigm. And what's more, he proved that that paradigm could be commercially successful. Tolkien did not invent secondary-world fantasy, but he is certainly the reason it is a marketing category today. Secondary-world fantasy, which in Tolkien's hands is as quirky and unpredictable and idiosyncratic as the work of any of his contemporaries, became popularized. With the advent of D&D, it became codified. And so imitating Tolkien's work today means imitating a horde of other people who've had the same bright idea between 1955 and now. It's not Tolkien's fault that the world he built has become cliché, but there's also no point in denying that it has. Diana Wynne Jones's Tough Guide to Fantasyland is proof enough of that.

So don't dance to Tolkien's drumbeat. Make up your own. Where you want to imitate Tolkien is in your follow-through, in building a world as rich and deep as you can.

4. MECHANICS

Roger Ebert said about Return of the Jedi: "The camera in 'Jedi' slides casually past forms of life that would provide the centerpiece for lesser movies." And while goodness knows I wouldn't cite the Star Wars movies as a great example of profound and brilliant world-building, it is absolutely true that one of the things the first trilogy does exceptionally well is that sense of what you might call depth of weirdness. In The Empire Strikes Back, Our Heroes reach the cloud city of Bespin. Why is there a city in the clouds? How does it stay up there? We don't need to know, and so we aren't told. The important thing is whether Lando is going to betray Han, and that's what we focus on. But in the meantime, we're in a city in the clouds, and it's beautiful and ridiculous and breath-taking. This kind of world-building, the kind that's built out of throwaway lines and marginalia, paradoxically by its very ellipticality, is the kind of world-building that can most convincingly create a sense of a secondary world as existing in four dimensions, as having heft and breadth and history.

World-building, to my mind, is something that a text does AS WELL AS telling a story. In the interstices of telling a story. In the background, in the margins, in the odd shadowed corners. Which isn't to say that the narrative can't stop for a while and settle down to explaining things, as, for example, the narrative of The Left Hand of Darkness is wont to do. But the reason world-building is clever and difficult, and the reason that speculative fiction writers talk about it a great deal, is that it's there in service of the story, not for its own further aggrandizement. The world-building in The Left Hand of Darkness is part of the story. The world-building in, for example, C. J. Cherryh's Rider at the Gate isn't, although the story is predicated on the world-building, and so Cherryh's world-building is always backgrounded, always oblique. But still there.

But please note that "infodump" and "exposition" are not the same thing. Beginning writers are told that infodumps are a terrible sin, and this makes them afraid of exposition and description. But exposition is not infodump, and infodump is not exposition. Robin McKinley's Spindle's End begins with a lovely and unapologetic piece of exposition; it's not an infodump, and that is because the exposition is also the story. Because story is not the same as plot.

I think of it as each novel having an apparatus. The apparatus has the calendrical systems worked out in loving detail, family trees, timelines, all the grunt work necessary to get the cogs to mesh and the machine to run. But the apparatus is not the novel, should not appear in the novel, and should not be necessary for the novel to be understood. In other words, the reader should be able to understand and enjoy the story without ever once looking at the map.

I was a bright and overachieving twelve-year-old; I understood the forms of fantasy and world-building, and I could mimic them. But I didn't understand what those forms were for. I didn't know why world-building mattered; I just knew you had to have a map.

Now, as a fantasist, as someone who reads, writes, and studies fantasy novels, I'd say you don't have to have a map.

But you do have to have a world.