Fantasy and Sci-Fi World-Building

Brandon Sanderson formalized principles for fantastical systems in a series of blog posts (2007–2013) that have become standard craft reference in the field. They apply to any fantastical element, not only magic.

The First Law: the ability to solve plot problems with a fantastical system is proportional to how well the reader understands it. Mysterious, barely-glimpsed magic can’t be used to resolve a conflict — the reader hasn’t been given the tools to evaluate whether the solution is earned. Magic the reader understands can resolve plot problems because the reader can follow the logical chain. The implication: if your climax depends on the magic system, your magic system needs to be comprehensible before the climax arrives.

This is a corollary of Internal Consistency: the reader can only evaluate whether the rules were followed if they know what the rules are. A writer who resolves the climax through a magic capability the reader didn’t know existed has violated the first law — and the reader will feel cheated even if they can’t identify why.

The Second Law: limitations are more interesting than powers. What the magic can’t do defines it as sharply as what it can do. An all-powerful system creates no stakes. A system with specific, meaningful costs and limitations creates drama at the point of limitation — the moment when the character needs to do something the system can’t (or shouldn’t) do.

The Third Law: expand what you have before adding new elements. Each new element a reader must track increases cognitive load and dilutes investment in existing elements. The writer who adds a new magical ability in the final act instead of developing an established one is trading investment for novelty — a bad exchange. The reader who has spent three hundred pages understanding one system and has begun to think creatively within it will be more engaged by that system under new pressure than by a new system arriving without the same investment foundation.

Sanderson’s own Mistborn trilogy demonstrates all three laws. The allomantic system — metals swallowed and burned to produce specific powers — is fully explained in the first act. Its limitations (finite supply, specific capabilities per metal, the cost of burning) are as important as its powers. When new capabilities are discovered, they arise from within the existing system’s logic rather than from outside it.

Any power without cost is dramatically inert. This is the practical content of Sanderson’s Second Law. Cost creates stakes at the moment of use. If using the magic requires something from the character — time, health, memory, years of life, moral compromise, relationships — then every use of the magic is a decision with consequences. The reader cares about that decision.

The cost must be specific and it must matter. Vague costs ("it tires her") accumulate into meaninglessness. Specific costs ("every use costs a memory, random and unrecoverable") create ongoing dread that the reader carries into every subsequent scene involving magic. The reader is now tracking the cost as a resource — one that’s being depleted in ways the character may not fully appreciate. The dramatic tension of the magic system is no longer just whether it works; it’s what using it is taking away.

The best costs are ones that threaten the character’s identity, relationships, or values — not just their physical capacity. N.K. Jemisin’s orogeny in the Broken Earth trilogy costs the orogene their connection to other people; the power that could save a city isolates the person who wields it. The cost is thematically integrated with the story’s argument about systemic oppression and survival. The magic system is not separate from the thematic premise; it expresses it.

See Magic and Technology Systems for extended coverage of system design, including the difference between hard magic systems (fully defined rules) and soft magic systems (atmospheric and undefined), and how the choice between them determines what kinds of plots become available.

Darko Suvin introduced the term "novum" (in Metamorphoses of Science Fiction, 1979) to describe the central estrangement device in science fiction — the new element that makes the world different from ours. The novum can be a technology (faster-than-light travel), a biological fact (a species with no fixed sex), a historical divergence (the Confederacy won), or a scientific discovery (contact with extraterrestrial intelligence).

The critical principle: the novum must have consequences that ripple through the society. Faster-than-light travel doesn’t just change how people travel — it changes colonialism, it changes loneliness, it changes war, it changes what words like "home" and "far away" mean. If the novum is limited to its direct effects without secondary and tertiary social consequences, the world-building is shallow.

This is the distinguishing mark of serious science fiction world-building. Competent SF establishes the novum and explores its first-order effects. The best SF thinks through to the second and third-order effects — the consequences of consequences — and builds a society that has been genuinely reshaped by them. When Le Guin imagines a world without gender, she doesn’t just remove gendered pronouns; she rebuilds the entire structure of political power, economic organization, family structure, and the concept of illegitimacy, because all of those things in our world are constructed around gender difference.

Ursula K. Le Guin’s Gethen (in The Left Hand of Darkness) demonstrates this fully. The novum is biological: the inhabitants of the planet Gethen have no fixed sex; they enter a periodic reproductive state and then return to a sexless default. Le Guin explores what this means for every dimension of social organization — there is no gender-based labor division, there is no gender-based power asymmetry, there is no concept of illegitimacy, there are no gendered pronouns. The novum generates an entire world that is genuinely different from ours in ways that illuminate what gender does in ours.

Ted Chiang’s Story of Your Life (filmed as Arrival) takes the novum of alien language and follows it to its second-order effect: what would it mean to perceive time differently? The alien Heptapod language encodes a non-linear experience of time; Chiang explores what a human being would become after learning to perceive the world that way. The story is not about the alien language. It’s about what the alien language does to a human mind, which is what all good novum-based SF is about: not the strange thing, but what the strange thing reveals about being human.

The best speculative world-building is not just setting — it’s thesis. The world is the thought experiment; the story is the examination of what the thought experiment implies. Le Guin’s Gethen argues about gender. 1984 argues about power and language. Kim Stanley Robinson’s The Ministry for the Future argues about climate economics and political will.

When the world-building is doing this work, every detail earns its place twice: as setting and as argument. The test for whether a world-building element is doing this second kind of work: does it complicate or illuminate the novel’s central question? If the answer is yes, it belongs. If it’s simply an interesting detail about the world, it faces the ordinary relevance test of all world-building.

The narrative argument and the world-building are not separate projects in speculative fiction at this level. They’re the same project approached from different directions. Building a world with certain specific rules forces certain specific questions; those questions are the novel’s questions; the story is the method of examining them. This is why Le Guin said that science fiction is not about the future but about the present: the novum defamiliarizes, which makes visible what ordinarily can’t be seen because it’s too familiar.

Fantasy can work the same way, though the tradition is different. Tolkien’s Middle-earth argues for the importance of ordinary courage and the corruption of power. A Song of Ice and Fire argues against the fantasy tradition of noble aristocracy — its world is structured to systematically punish the assumptions of genre romance. N.K. Jemisin’s Broken Earth builds an entire planetary geological system and social caste structure specifically to examine the logic of systemic oppression. In each case, the world-building and the argument are inseparable.

The information delivery challenge in speculative fiction is acute. Readers need to understand the rules, but delivering rules in blocks kills pacing. The solutions are limited:

Introduce rules as they become relevant. The character encounters the limitation at the moment the plot requires the limitation. The reader learns the cost structure when a character is forced to pay a cost. The rule arrives in scene, under pressure, which makes it memorable and contextually grounded rather than filed away as background information.

Embed rules in character conflict. A disagreement between characters about how the magic works, or should be used, delivers rule information as dramatic content. The reader is learning about the system while watching people argue — which is more engaging than being told about the system directly.

Use the newcomer. The character who doesn’t know the rules asks the questions readers would ask. The fish-out-of-water structure delivers exposition through genuine dramatic need: the character needs to know this to survive, and the reader is learning alongside them. See Science Fiction 1a — The World as It Appears for the SF-specific approach to establishing the baseline before the novum disrupts it — the baseline work that makes everything that follows legible.

Demonstrate before explaining. Show the magic operating before explaining how it operates. Readers form initial models from demonstration; explanation refines those models rather than building them from scratch. Explanation before demonstration produces rules that feel abstract; demonstration before explanation produces rules that feel discovered.

The worst approach is the information dump: several paragraphs or pages of expository explanation before the story’s events make any demand on the reader to understand the system. This is the failure mode of writers who have done their world-building work and want to share it — who treat the reader’s understanding of the system as a precondition for the story rather than something the story itself can develop. The reader doesn’t need to understand the complete system before the story begins. They need to understand as much as each scene requires, and no more.