← Files NovelistARCHIVED FILE
skills/novel-story-telling/references/mystery.md
6.93 KB · Oct 5, 2026 · 18:29 UTC
# Mystery And Detective Story Design ## Contents 1. Reader contract 2. Build backward from truth 3. Recommended flow 4. Clues and reveals 5. Escalation and resolution 6. Continuity ledgers 7. Failure modes 8. Research basis ## Reader Contract Treat the mystery as an intellectual and emotional promise: the solution must become more inevitable as it becomes more surprising. Give the reader access to decisive observations before the solution, while allowing characters to misinterpret their significance. Use historical fair-play rules as principles rather than rigid prohibitions. The central requirements are causal truth, available evidence, established culprit or cause, and a solution earned through inference rather than accident. ## Build Backward From Truth Write the immutable hidden account before outlining investigation chapters: 1. What happened? 2. What exact sequence and timing made it possible? 3. Who acted, assisted, witnessed, lied, or misunderstood? 4. What did each person perceive and believe? 5. What physical, digital, social, and testimonial traces must exist? 6. Which traces permit a plausible wrong interpretation? 7. Which combination makes the final inference stronger than alternatives? 8. How did the responsible party conceal, stage, erase, or exploit those traces? Do not change this truth casually during drafting. If revision changes it, recompute every clue, alibi, testimony, action, and knowledge state downstream. ## Recommended Flow 1. **Disruption:** present the crime, disappearance, impossible event, accusation, or anomaly. 2. **Investigation contract:** establish the central question, stakes, investigator limits, and initial suspect or cause space. 3. **First cycle:** observe → hypothesize → test → gain fact → open sharper question. 4. **Differentiation:** give each credible suspect motive, opportunity, concealment, and a non-central reason to lie. 5. **Model failure:** invalidate a promising explanation because a real clue was interpreted wrongly. 6. **Midpoint reframe:** change the governing question through wrong victim, wrong time, staged location, concealed relationship, or two events mistaken for one. 7. **Compression:** narrow alibis and escape routes; add a deadline, institutional pressure, or threat to evidence. 8. **Proof chain:** connect motive, means, opportunity, action, concealment, and contradiction. 9. **Reveal and test:** present causal order; use reconstruction, trap, experiment, or independently verifiable consequence when appropriate. 10. **Aftermath:** resolve legal, relational, and epistemic consequences, including meaningful red herrings. ## Clues And Reveals Give every decisive clue: - an exact observation; - an exposure chapter; - a surface interpretation; - a true interpretation; - a corroborating fact; - a reframe or payoff chapter. Use the sequence `show → misclassify → stress → reconnect → prove`. Track separately: - what objectively happened; - what the investigator knows; - what the reader has seen; - what each suspect knows; - what each suspect believes the investigator knows. Hide significance more often than existence. Do not suppress an immediately relevant viewpoint thought solely to preserve surprise. Red herrings require a cause: unrelated wrongdoing, shame, protection of another person, mistaken memory, institutional incentive, ambiguous procedure, or deliberate staging. Resolve their causal meaning even when they do not identify the culprit. ### Investigation scene contract - question entering the scene; - method: inspect, interview, reconstruct, research, surveil, test, or confront; - obstacle or deceptive frame; - new observation; - inference or hypothesis change; - suspect, relationship, or risk change; - next actionable question. Combine scenes that produce no fact, inference, cost, or changed relationship. ## Escalation Devices - A witness changes testimony to protect a different secret. - Genuine evidence is staged in the wrong context. - Two suspects share an alibi while protecting different acts. - Exposing private wrongdoing costs the investigator cooperation. - A deadline threatens evidence, another victim, trial, inheritance, election, or departure. - The culprit adapts after learning the current hypothesis. - Solving one impossibility reveals a larger one. - An institution rewards a convenient suspect or explanation. - A technically correct finding carries a socially false implication. - The investigator must choose between trust and a necessary test. Escalate through narrowing possibilities, rising consequence, adversarial adaptation, and emotional cost, not by adding arbitrary bodies. ## Resolution Devices - **Inference chain:** combine several ordinary observations into one unique causal account. - **Reconstruction:** reproduce the mechanism under controlled conditions. - **Contradiction:** demonstrate that one claim cannot coexist with a verified fact. - **Behavioral trap:** create a situation where only someone with hidden knowledge reacts meaningfully. - **Independent consequence:** predict a trace or event that the true model alone explains. - **Moral reveal:** show why people lied as well as what happened. Distinguish suspicion from proof. Explain identity, method, motive, opportunity, concealment, and the role of major red herrings. ## Continuity Ledgers ### Clue ledger ```yaml clue_id: observation: source: reliability: found_at: observers: [] reader_exposure: surface_inference: true_inference: corroborators: [] payoff: status: unseen | shown | misread | reframed | resolved ``` ### Suspect matrix Track motive, means, opportunity, alibi claim, verified alibi parts, lies and reasons, unrelated secrets, victim relationship, evidence knowledge, investigator belief, and reader-facing suspicion. ### Truth and reported timelines Keep immutable event rows with bounded time, actor, action, location, witnesses, evidence created, and concealment. Keep claimed accounts separately so contradictions remain visible. ## Failure Modes - Decisive evidence appears only during the solution. - The culprit or true cause is introduced too late. - Red herrings have no causal explanation. - Characters lie only because delay is needed. - Forensic, magical, or technological rules shift for the reveal. - The investigator guesses without a proof chain. - Interviews repeat exposition without changing a hypothesis. - Surprise depends on withheld viewpoint knowledge. - The reveal identifies the culprit but not method, motive, opportunity, or concealment. - Revision changes the truth without propagating effects through the clue ledger. ## Research Basis - S. S. Van Dine, [Twenty Rules for Writing Detective Stories](https://jhnet.sakura.ne.jp/petit/20rule_e.html). - Writing Excuses, [Building a Mystery](https://writingexcuses.com/18-08-building-a-mystery/). Use dated prescriptions critically. The workflow above preserves fair evidence and causal reasoning while allowing modern forms, multiple investigators, romance, non-murder mysteries, and hybrid genres.
SHA-256: 9345b363ca7bf23572e8e5fdbcfef9d4ee23135c66b59bbd66f298a424b013f2