Self-Consistent Lore

Abstract

Problem: Developers and players often conflate "realism" with "consistency," leading to dismissed feedback and confused expectations.

Approach: Tim Cain distinguishes realism (mimicking real-world features) from self-consistency (establishing world rules and never breaking them), drawing on his experience designing Arcanum and other RPGs.

Findings: Self-consistency is the foundational contract between designer and player. Unreliable narrators and mystery are powerful tools but must be used sparingly β€” too much makes lore feel random. Lore drift across sequels is inevitable but must be managed carefully.

Key insight: Your game world doesn't need to be realistic β€” it needs to follow its own rules. Consistency is the bedrock; mystery and unreliable narration are spices that enhance it but can ruin it in excess.

Source: https://www.youtube.com/watch?v=1MCZ-AFbAAs

Realism vs. Self-Consistency

Tim opens by drawing a sharp line between two concepts that players, reviewers, and even developers constantly conflate:

  • Realism is about mimicking real-world features β€” weather, gravity, time of day, flight physics. Even when imperfect (random weather, exaggerated jumping), it's still an attempt to simulate reality.
  • Self-consistency is about establishing rules for your world and then never breaking them. It has nothing to do with whether those rules match our world.

The problem arises when developers dismiss player complaints with "we're not trying to be realistic" when the actual complaint is that the game's own lore contradicts itself.

The Arcanum Example

Tim's go-to example of self-consistency comes from Arcanum's earliest design sessions β€” just him, Leonard Boyarsky, and Jason Anderson in a living room:

They established one foundational rule: magic and technology are antagonistic. Magic suppresses natural law in the mage's vicinity; technology uses and reinforces natural law. This meant:

  • Spells weakened around tech items
  • Tech devices failed around powerful mages
  • The player had a personal magic/tech rating with measurable, concrete effects on spells, skills, and items

Everything in the game obeyed this rule. That's self-consistency β€” a designer deciding something is law and enforcing it across every system.

The Unreliable Narrator

A critical point: self-consistency is not the same as NPC dialogue being accurate. You cannot point to an NPC's words and claim the lore is broken, for two reasons:

  1. NPCs lie. They have agendas. The designer wrote the lie deliberately. If you don't catch it, that's on you.
  2. NPCs are wrong. They believe what they're saying but are simply mistaken β€” the classic unreliable narrator.

Tim's example: an NPC tells you there's a fire-breathing monster in a cave. In reality, it's just a bear. Someone got mauled, the story got exaggerated over time, and now the village thinks there's a dragon. The NPC believes every word. They're still wrong.

The Locked Creature Thought Experiment

If you show a creature locked up without food or water for a hundred years, you've implicitly stated that creature doesn't need food or water. Unless:

  • There was a food source for 99.5 of those years
  • The "hundred years" is an unreliable narrator's exaggeration β€” maybe it's been two days
  • There's a hibernation mechanism, cryo-chamber, or time dilation at play

The designer may know which explanation is true. They don't have to reveal it. This connects to Tim's philosophy on leaving mysteries β€” unexplained details invite multiple player interpretations, all stemming from one consistent canonical truth. That ambiguity keeps players engaged and debating.

Mystery as a Double-Edged Sword

Mysteries and unreliable narration are powerful tools, but Tim warns: too much will destroy your lore. If players encounter excessive mystery, conflicting narrators, and unexplained contradictions, the world starts feeling random β€” as if there's no lore at all and the designers are making it up as they go.

Tim's firm position: decide your setting, then your story, then build system mechanics that support both. Don't improvise your foundations.

Lore Drift

Lore drift β€” the gradual shifting of established lore across sequels, spin-offs, and studio changes β€” is inevitable. Some of it is intentional (retroactively declaring a previous narrator unreliable). Some is structural (a sequel narrows the possibility space the original left open).

But excessive, careless drift will lose players, just like excessive mystery will.

The Spice Analogy

Tim frames all of these tools β€” unreliable narration, mystery, lore drift β€” as spices:

"A little makes a dish nicer. Too much will ruin it."

How much is too much? Every designer has a different tolerance, and so does every player. There's no universal number. But somewhere along the continuum, you'll start losing people β€” and that's the line developers need to feel for.