This post is part two of two in a series on worldbuilding. To read part one, click here.
Last week, we explored how to plan and develop a speculative world, how we must situate ourselves along a continuum between the real and the absurd, how we can tie into existing cultural concepts while still making our worlds unique, how worldbuilding can function as social critique, worlds’ inner consistency, and their service to the stories they contain.
Today, let us wander forward into those stories themselves. How does a constructed world manifest in the text itself? How can we best guide readers through it, and what techniques may be better avoided?
We’ll look at a sci-fi example along the way, explore alternatives, and hopefully, by the end, have some clearer ideas about how to make these worlds come alive.
Info-dumping: the joy and the peril
In the most famous speculative worlds of the mid twentieth century (I’m thinking 1984‘s Oceania and Brave New World‘s hatcheries, and of course Tolkien’s Middle Earth) worldbuilding was detailed, systematic, often club-you-over-the-head thorough. Brave New World opens with a room-by-room tour of the baby-making facility in all its clinical precision. It is tacitly embedded into the actual story by casting the tour as an orientation for future hatchery managers. 1984 waits a while longer for its full grand, explicit world-build, but when Winston and Julia sit down to read Emmanuel Goldstein’s book, readers get a detailed, textbook-style description of the world’s political structure. The novels of Middle Earth are famous for their detailed cultural, military, and linguistic worldbuilding.
The technique these three examples use we can today call info-dumping, in which copious details flesh out the world in the text. Some people do love this, and there is absolutely a certain fascination with seeing how far these worlds can go. In the high school Creative Writing Club I supervised in Minnesota, students took great joy in crafting these worlds and displaying them out for others. It is a kind of immersive escape, a baroque ornamentalism by the brain. In high fantasy still, this kind of worldbuilding holds some currency. These worlds create a suspension of disbelief, where the questions all have answers, where every object has a history and every word a significance.
And yet, I think the perils of this explicit, didactic kind of artistry are great. For every reader enthralled in 1984, there are a dozen who find it a slog. I myself as a teenager made it halfway through The Return of the King, twice. I couldn’t sustain myself through one of the fifty-page battles, I think.
This is the info-dump’s more prosaic side, notorious in online writing advice as an interest-killer for readers. I tend to agree. Detailed explanations of our worlds, even as they trace the proliferation of our imaginations, often stymie the readers’.
A lighter hand: the power of the guided inference
Lincoln Michel in his somewhat controversial article “Against Worldbuilding” proposes the alternative of worldconjuring. “Worldconjuring uses hints and literary magic to create the illusion of a world,” Michel says, “with the reader working to fill in the gaps. Worldbuilding imposes, worldconjuring collaborates.”
I like this idea, and it expresses cleanly what I think many of us world-builders ultimately want anyway. The idea is that the world does not exist in the story to be transferred to the reader; rather, the story’s world develops through interaction between the text and the reader’s mind.
Let’s start with an example. Here’s an excerpt from David Mitchell’s Cloud Atlas, early in the section “An Orison of Sonmi-451”:
Fabricants have no earliest memories, Archivist. One twenty-four-hour cycle in Papa Song's is indistinguishable from any other.
Then why not describe this "cycle"?
If you wish. A server is woken at hour four-thirty by stimulin in the airflow, then yellow-up in our dormroom. After a minute in the hygiener and steamer, we put on fresh uniforms before filing into the restaurant. Our seer and aides gather us around Papa's Plinth for Matins, we recite the Six Catechisms, then our beloved Logoman appears and delivers his Sermon. At hour five we man our tellers around the Hub, ready for the elevator to bring the new day's first consumers. For the following nineteen hours we greet diners, input orders, tray food, vend drinks, upstock condiments, wipe tables, and bin garbage. Vespers follow cleaning, then we imbibe one Soapsac in the dormroom. That is the blueprint of every unvarying day.
This is weird. When I first read this, I remember my mind racing. There’s a lot of information here, some gesturing at religious services, then at fast food dining. Yet the description is not given for the benefit of an unfamiliar reader (although of course, ultimately, it is), but rather it is the narrator’s description of her daily life, and just as you and I would never step out of our own minds to describe what “coffee” or “brushing teeth” are to an alien observer, so the narrator has no reason to explain “yellow-up” or “Papa’s Plinth” or the “Soapsac.”
So although a lot of information is indeed presented, at least of equal importance is what has been omitted, and that is where we readers work feverishly to fill in the gaps.
This worldconjuring I see as a type of guided inference: Mitchell has been careful to give us enough that he is confident we will fill in the gaps in the way he intends. If he gives us too much (an info-dump), we become passive receivers, less involved, less invested. If he gives too little, we become lost, confused, and lose interest. This type of writing requires careful attention to the mind of the reader, thought experiments of the how-would-I-interpret-this-if-I-didn’t-know-the-story type, and likely consultation with beta-readers and a fair amount of tinkering beyond the first draft.
Info-dumping harms reader engagement because it insists on filling in every detail. To use an analogy from classroom teaching, info-dump worldbuilding is a teacher-driven lecture. Worldconjuring, then, is a more student-centered, inquiry-based lesson.
It’s worth noting that, just as a lecture does ultimately convey more total information than inquiry-based learning, the worlds that develop in the reader’s mind are probably not as thorough, detailed, or even the same as what the writer intended. And with this we, as writers, must reckon.
Specific tools for guided inference
Looking back at the Cloud Atlas excerpt above, let’s see what tools Mitchell is using to conjure this world.
Word choice: repurposed vocabulary
One of the first techniques we can see is the use of familiar vocabulary in new contexts. That is to say, for the new concepts of his world, Mitchell employs recognizable vocabulary rather than made-up words that would remain stubbornly opaque.
The narrator first calls herself a “fabricant.” Of course, new to this world, we can’t know precisely what is being described. However, we know what it means to fabricate something. We can make an educated guess that this narrator is some kind of clone or robot.
The next new term is “stimulin,” and this is even easier: the low-hanging fruit, perhaps–it’s easy to imagine someone being awoken by a stimulant. So we can figure out what this is, and with these two terms we can start to assemble a clearer inference: this narrator is not like most humans; rather, she has been made artificially, and she is minutely controlled, down to the time of her waking up.
“Yellow-up” is more obscure. Worth noting is that the general idea of this sentence has already been conveyed with the clearer “stimulin,” and so we needn’t necessarily understand “yellow-up” to get the basic picture. For me, the relative opacity of this term helped intrigue me–the world felt more mysterious and I wanted to give more thought to it. As a note, I think a couple of interpretations of “yellow-up” are probably reasonable. I first thought of it as in the video game term “power up,” which would suggest another drug like the stimulin. Now, however, I’m thinking that the “yellow-up” describes a simulated sunrise of some kind, maybe displayed on the wall.
Ultimately, Mitchell’s word choice provides one of the clearest roads in for readers to access, decipher, and construct the world according to the author’s specifications. Perhaps it is we who are the fabricants. These words are the stimulin. Or maybe that’s taking it a bit far.
In media res
A more holistic technique used in the above passage is the unceremonious throwing-headfirst of the reader into the world. We can liken this to the common advice about plot development, that rather than beginning at the beginning, a story’s narrative should begin with the action, in the middle, in media res.
In worldbuilding, we achieve this by not explaining (at least not at first) the origin or the reason for any of the world’s elements. Rather, we show them as the characters who inhabit this world experience them. They are normal. It is we, the reader, who are the outsiders, and this causes the reader to scramble to interpret, thus fully engaging right away in the world.
There’s absolutely a sense of this scrambling as Sonmi describes her everyday life, and we readers must rapidly adjust our sense of the world in order to correctly interpret her description. And within just a few paragraphs, a monstrous picture has been conjured, of cloned fast-food workers ruthlessly controlled, all for a shamefully profit-driven master.
The risk here, certainly, is that readers become confused. It’s a fine line to tread. We must provide enough clues in order to help the reader build a cohesive world, yet if we give too much, we are back in lecture territory.
World through character
The third technique I see Mitchell using here is a subtler one still. It is an indirect form of worldcrafting, and in some ways, I think, the most powerful and authentic: it is the painting of this world through the characters it has spawned.
Emerging from the assumption that a person’s surroundings influence their identity, we see Sonmi’s world most authentically not in the surroundings themselves, but in Sonmi herself. Sonmi’s first statement, that “Fabricants have no earliest memories,” suggests immediately a kind of plundered humanity. Something in the character’s world has robbed her of basic human freedoms like memory, and later we hear that she has been robbed too of the autonomy of waking up at a time she chooses, of basic curiosity, and her use of capital letters on words like “Six Catechisms” suggests a brainwashed religiosity towards the fast-food restaurant for whose profit she exists.
It is easier to believe such a world when we see the effects it has on its inhabitants. In our writing, we can consider how the societies we develop mold our characters. Animating those characters is a window to that world.
A world is a sea; the story is a funnel
In order to guide readers to the inferences we want, we must know our worlds well. We might build detailed worlds in our own minds, knowing that most of what we imagine will never make it onto the page. Yet it is because our readers are actively rebuilding these worlds in their own minds that we must consider all of the implications of what we write. If we write about Papa Song’s restaurant without, either before the first draft or after, considering the wider context in which it exists, we risk our readers making a logical leap that breaks their suspension of disbelief.
The information that explicitly ends up in the story is like the small end of a funnel. We may mourn all of the worldbuilding we did in our own minds when we see how little of it spills out onto the page. But ultimately, it is there still. The funnel leads back up into the whole sea of the world, built anew in the reader’s imagination. We must release control of our stories a little bit. We must let the readers be the artists too.
Worldbuilding, worldconjuring, these are big topics. How do you handle them? What challenges have you faced? What tools do you use? What other worlds in literature have you especially enjoyed and why? I’d love to hear. Thanks, and best wishes,