TL;DR
Build worlds that hold up when readers push on them. Most writers build backward from a strong image and watch it collapse at the first hard question. Build from function outward instead, starting with survival and following the logic, and a whole culture assembles itself around honest constraints. The course also shows how to use AI as an assistant to stress-test your world’s logic, never to invent the world for you.
Most writers build worlds backward. They start with the strong image, the floating city, the underground kingdom, the society powered by crystals, and then try to reverse-engineer an explanation for how the thing could possibly work. It looks great in concept art and falls apart the moment a reader asks a simple question. How do they eat? Where does the water come from? The illusion collapses because there was never anything holding it up.
A floating city with no visible food supply makes me a little crazy. The concept art looks gorgeous, and then one question about dinner sinks it. What gets me is how avoidable it is. Ten minutes spent asking how these people eat would have saved the book.
This course teaches the opposite. You build from function outward, starting with survival and expanding through logical necessity, so the world holds together no matter how hard a reader pushes on it. The strong images still come. They just sit on a foundation that can bear weight.
Every civilization solves the same core problems
Start with survival and the culture assembles itself.
Strip away the setting and every society that ever worked had to solve the same handful of problems. Get food. Manage water. Provide shelter. Keep people safe. Coordinate group behavior so the world doesn’t fly apart. How your invented society answers those questions determines everything else, the culture, the technology, the politics, the feel of an ordinary day. Start there and the rest grows on its own. Start with the crystal towers and you’re always patching holes.
The discipline is to ask the boring questions first and follow the answers wherever they lead. A society’s values aren’t something you assign. They fall out of how that society survives. Get the survival logic right and the culture writes itself, consistent all the way down because it all traces back to the same root.
I’m partial to boring questions, and I trust a writer more for asking them. Food and water sound dull next to crystal towers, but they’re where the good material comes from. A culture that grew out of how people survive feels lived in. A culture assigned by the author feels like a theme park.
Watch: Lecture #5: Worldbuilding Part One — Brandon Sanderson on Writing Science Fiction and Fantasy, Brandon Sanderson
The water problem
Take water, because it shows the method better than anything.
Water availability doesn’t merely influence where people settle. It determines social structure, property law, politics, and values. A desert society doesn’t simply value water more than a river society. It develops an entirely different approach to ownership, cooperation, and conflict, because survival depends on managing a resource that people in wet climates never have to think about.
Follow it. In a water-scarce world, private hoarding kills the community, so individual ownership of water becomes almost meaningless and sharing becomes a survival rule enforced by everyone. That single fact ripples outward. It shapes who holds power, what counts as the worst crime, how strangers are treated, what the religion sanctifies. You didn’t decide any of that. You decided the world was dry and followed the logic, and a whole culture assembled itself around one honest constraint. That’s functional world-building. Pick the constraint, then obey it.
The water example sells me on the whole method, because it shows how much story falls out of a single honest decision.
You didn’t have to invent a religion or a legal code by hand. Make the world dry, follow it, and those show up on their own. That’s a far better deal than building everything piece by piece.
Settlement follows resources
Where people live is never random, and inventing a map before you know why anyone would settle each spot is building backward again. People cluster where the core problems are easiest to solve, near water, near food, near defensible ground, near trade routes that bring what the local land cannot. A city exists because that location solved a problem better than the alternatives. Know the problem it solved and you know the city’s character, its wealth, its anxieties, what it fears losing.
This turns your map into a set of answers instead of a set of shapes.
Every settlement is there for a reason, and the reason tells you how the people think. A town built around a single well behaves differently from a port that touches the whole world. You get variety for free, because different constraints produce different people, and none of it feels arbitrary because none of it is.
Random maps annoy me. You can spot them right away, a port city with no trade or a capital on a bare hill for no reason. Every settlement should answer a question, and if you can’t say which question, move it or cut it.
Space carries meaning
People attach meaning to place, and your world should too.
Humans mark territory, sanctify some ground and profane other ground, and anchor memory to physical space. A threshold means something. A seat at the head of the table means something. The room nobody enters means something. When your characters move through the world, the world should push back with these weights, so that walking into a particular building is walking into everything that building represents to them.
This is where a functionally sound world becomes an emotionally real one.
The logic makes it believable. The meaning attached to its spaces makes it matter. A reader should feel that your characters belong to their world, shaped by its constraints and carrying its meanings, so the setting isn’t a backdrop they stand in front of but a force that made them who they are.
This is the part most world-builders skip, and it’s the part I care about most. A world that works mechanically but carries no meaning feels like a very good spreadsheet. Give a place weight in a character’s memory and the reader suddenly cares when it burns.
How can AI help with world building?
The world is yours to invent. The constraints you pick and the logic you follow are the craft. But a tool is useful for the checking, and that’s where invented worlds break. You can use AI to hunt a manuscript for the moment your own rules contradict each other, the well that runs dry in one chapter and overflows in the next. You can have it pressure-test a society by asking the simple survival questions you might have skipped, so the holes surface before a reader finds them.
You can use it to explore the downstream consequences of a constraint, then keep the ones that fit and discard the rest. It stress-tests the logic. You build the world. Build from function and a world will survive any question a reader throws at it. Build from a pretty picture and you’ll spend the whole book patching holes, and readers find those holes faster than you’d think. Give me a plain world that holds together over a dazzling one that falls apart by page forty, every time.
Explore the hub
Writing Courses: the full set of writing courses, gathered in one place.