“An empire where memories can be taxed” is not yet a world. It is a spark. The difference between a clever sentence and a usable core premise is whether the idea can reliably generate characters, conflicts, institutions, scenes and consequences without needing a new invention every five pages.
This worked example shows how to turn that rough spark into production-ready material. The goal is not to produce a giant encyclopedia. It is to make a small set of decisions that other parts of the story can lean on.
Brandon Sanderson’s public BYU worldbuilding lectures repeatedly return to a practical distinction: a creator cannot explore everything at equal depth, so the work benefits from choosing where detail will matter and how rules create consequences. That principle is useful even if your style is nothing like Sanderson’s. A core premise becomes strong when it tells you what to leave out as clearly as what to include.
The rough idea
Start with one sentence:
In the river-city of Lumen, citizens pay part of their annual tax by surrendering certified memories to the state archive.
It has novelty, but almost no operating logic. We do not know whether memories are copied or removed, who values them, why a government wants them, how fraud works, what happens to poor citizens, or what kind of story this system creates.
Before adding lore, ask five questions.
1. What human pressure does the premise create?
If memory tax is only colorful bureaucracy, it will decorate scenes rather than drive them. Give it a pressure people understand without exposition.
Decision: surrendered memories are removed from the citizen, not copied. The state grades them by emotional intensity, strategic value and rarity. A person can pay with many ordinary memories or one highly valued memory.
Now the premise creates sacrifice. A parent might pay with the memory of a dead spouse to keep the family business. A veteran may possess politically dangerous memories that are worth too much. Wealthy people can purchase other citizens’ memories through licensed brokers rather than lose their own.
The rule has produced class conflict, family conflict and moral conflict in one move.
2. What does the system make cheaper, and what does it make expensive?
Every institution creates asymmetry. If you know what becomes cheap and what becomes scarce, the economy begins to emerge naturally.
Decision: low-grade memories are plentiful and cheap; verifiable firsthand memories of disasters, crimes, elite ceremonies or extinct places are scarce. Memory authentication becomes a serious profession. Counterfeit memories become a serious crime.
This creates occupations without inventing them arbitrarily: auditors, memory brokers, forensic examiners, illegal editors, archive clerks, trauma specialists and smugglers. It also creates physical places: certified extraction clinics, black-market salons, state vaults and neighborhood “memory cooperatives.”
Notice the method: we are not asking “what cool jobs exist?” We are asking “what jobs must exist if the premise is true?”
3. What can the system not do?
A premise with no limits becomes a universal explanation machine. That kills tension.
Decision: extracted memories cannot be played like perfect video. They are subjective sensory records. They can omit peripheral facts, contain false interpretations and degrade when repeatedly transferred. Editing leaves detectable traces, but detection is not perfect.
These limits preserve ordinary investigation. A murder memory is evidence, not instant truth. A ruler cannot simply download perfect expertise from a genius. A character can sincerely remember something incorrectly.
The limitation also protects the story from the common “why don’t they just use the magic?” problem.
4. Who benefits enough to defend the system?
Bad institutions survive because someone receives value, not because everyone is foolish.
Decision: the archive uses collected memories to train navigators, preserve endangered craft knowledge, reconstruct disasters and maintain a national historical record. Many citizens genuinely credit it with preventing repeated catastrophes. The ruling class also exploits the same archive for surveillance and political leverage.
Now the system cannot be reduced to “evil government steals memories.” Abolition would destroy real public goods; reform threatens entrenched power; doing nothing preserves injustice. That gives factions something meaningful to argue about.
5. What ordinary habit proves the premise is real?
Readers believe worlds through behavior. Add one daily practice that a person in this society would consider normal.
Decision: families keep annual “anchor books” — handwritten summaries of important events — because nobody knows which memories they may later surrender. Before tax season, people reread old anchor books together and decide what they are willing to lose.
That is more useful than another paragraph about the Ministry of Mnemonic Revenue. It gives you a scene.
Build the premise card
After those decisions, compress the world into a one-page reference that a writer, illustrator or collaborator can actually use.
| Field | Working answer |
|---|---|
| Core rule | Annual taxes can be paid with removed, authenticated memories |
| Scarce resource | rare firsthand memories with high evidentiary or historical value |
| Everyday consequence | families document experiences they may later forget |
| Main inequality | wealthy citizens can buy substitute memories |
| Institution | National Memory Archive and licensed extraction network |
| Black market | edited, stolen and falsely attributed memories |
| Hard limit | memories are subjective, degradable and imperfect evidence |
| Public benefit | disaster learning, craft preservation, historical record |
| Abuse vector | surveillance, coercion, political blackmail |
| Story question | what should a society be allowed to own about a person’s past? |
This table is not the story bible. It is the spine. Anything added later should either deepen one row, create a useful contradiction, or be cut.
Turn the premise into three character positions
A usable core premise should generate people who want different things for good reasons.
Mara, archive examiner. She believes the system saved the city after a flood because engineers learned from thousands of extracted eyewitness memories. Her job is to catch fraudulent submissions. She is not naive; she has seen corruption and thinks reform is possible.
Teren, memory broker. He legally buys memories from cash-poor citizens and sells them to wealthy clients who need tax credits. He tells himself that the market gives poor people choice. His clients call him useful. His neighborhood calls him a scavenger.
Ilye, dockworker and parent. She owes back taxes after an injury and has one memory valuable enough to clear the debt: the night she saw a senior official order a lethal gate closure during the flood. Paying the tax would erase the memory from her while putting it inside the institution she does not trust.
We now have a plot without inventing a villain. The premise itself pushes the characters together.
Test the premise with five scene prompts
If a core premise cannot generate scenes, it is still too abstract. Try these before writing chapters:
- A family dinner where everyone argues over which shared memory one person should surrender.
- A courtroom where two authentic memories of the same event contradict each other.
- A rich client buying a stranger’s childhood memory because it has unusually high tax value.
- A public museum exhibit assembled from memories whose original owners no longer remember the event.
- A black-market editor offering to remove the shame from a memory while keeping its official value.
Each prompt reveals a different dimension — family, law, class, culture, crime — without requiring a new magic system.
Run the “three attacks” test
Before promoting the premise into canon, attack it from three directions.
Economic attack: if memories have value, can people farm them? The answer might be yes, but only novel firsthand experience receives high grades, creating dangerous “experience tourism.” That is a consequence worth keeping.
Political attack: why would citizens accept the system? We already gave it public benefits and a long history after catastrophe. Add elections around tax rates and archive access. Acceptance becomes contested rather than implausible.
Plot attack: can a clever character solve every problem by extracting someone’s memory? No, because memory is subjective, access is regulated and extraction removes the memory from the source. Getting evidence has a cost.
If one attack destroys the premise, fix the rule now. Do not write 80,000 words hoping readers will ignore it.
What we deliberately did not design
We did not name twelve ministries. We did not draw the entire continent. We did not invent the archive’s architecture, religious doctrine, school curriculum or military ranks. Those may become necessary later, but the core premise does not need them yet.
That restraint is part of the work. Good worldbuilding is not measured by the amount of lore produced; it is measured by how much useful story pressure each decision creates.
The finished result can now support a novel chapter, a game quest, a comic arc or a pitch deck. More importantly, another creator can read the premise card and make compatible decisions without asking what the world is “supposed to feel like.” That is the point at which a rough idea becomes usable creative material.
Related Reading
- How strong creators design Core Premise: decisions hidden behind the finished work
- Browse all Lore Lab articles
Sources and craft references
- Brandon Sanderson — BYU 2020 Lecture #5: Worldbuilding Part One
- Brandon Sanderson — BYU 2020 Lecture #6: Worldbuilding Part Two
- University of Chicago Press — Writing Fiction, Tenth Edition
The Lumen setting, characters and memory-tax system in this article are original worked examples created for teaching the method; they are not historical claims.