diff --git a/docs/economy/RUMOR-SYSTEM-0001.md b/docs/economy/RUMOR-SYSTEM-0001.md
new file mode 100644
index 0000000..efaf45c
--- /dev/null
+++ b/docs/economy/RUMOR-SYSTEM-0001.md
@@ -0,0 +1,341 @@
+# RUMOR-SYSTEM-0001
+## Rumor, Information Delay, and Informal Markets
+### Status: Canonical Economy Seed
+### Layer: OTIVM (Roman Merchant)
+### Purpose: Define how uncertain information moves through the city, alters prices, creates opportunity, and misleads actors
+### Repository Path: docs/economy/RUMOR-SYSTEM-0001.md
+
+---
+
+## 0. Design Intent
+
+Markets do not wait for certainty.
+
+Most economic decisions are made before facts are known. Men act on smoke, absences, raised voices, late carts, unusual purchases, and guarded expressions.
+
+This document prevents a false simulation where all actors receive perfect information at the same time.
+
+In OTIVM:
+
+- rumor is an economic input
+- ignorance has geography
+- truth arrives unevenly
+- reputation alters belief
+- opportunity often exists only during confusion
+
+---
+
+## 1. Definition of Rumor
+
+Rumor is not merely falsehood.
+
+Rumor is socially transmitted, incomplete information with uncertain accuracy.
+
+It may be true, partly true, outdated, intentionally distorted, a misinterpreted signal, or speculation mistaken for fact.
+
+A rumor can move markets even when false.
+
+---
+
+## 2. Why Rumor Matters Economically
+
+Before official confirmation:
+
+- prices move
+- buyers hesitate
+- lenders tighten terms
+- hoarding begins
+- labour shifts districts
+- transport reroutes
+- opportunists buy quietly
+
+The first reaction is often based on rumor, not truth.
+
+---
+
+## 3. Primary Sources of Rumor in Ostia
+
+| Source Node | Typical Information |
+|---|---|
+| BALNEA | status shifts, meetings, disputes, political whispers |
+| riverfront | arriving cargo, delays, losses, foreign news |
+| HORREA | shortages, inventory pressure, distress selling |
+| workshops | fires, wages, tool scarcity, output decline |
+| taverns / cauponae | labour news, theft, street violence |
+| legal forum | debts, seizures, petitions, enforcement |
+| stables / yards | cart shortages, route condition, mule health |
+| households / servants | private affairs entering public speech |
+
+No source is perfectly reliable.
+
+---
+
+## 4. Information Classes
+
+| Class | Meaning |
+|---|---|
+| signal | directly observed fact, such as visible smoke |
+| report | witness says event occurred |
+| inference | likely consequence guessed from fact |
+| embellishment | dramatic additions |
+| agenda_rumor | shaped to benefit speaker |
+| denial | false minimization or concealment |
+
+Example:
+
+Smoke over forge = signal.  
+“The owner fled with debts” = rumor.
+
+---
+
+## 5. Core Parameters
+
+| Token | Type | Meaning |
+|---|---|---|
+| rumor_velocity | city | speed of spread |
+| rumor_accuracy | scenario | closeness to truth |
+| source_credibility | actor/relation | trustworthiness of speaker |
+| information_delay | relation | days before actor hears usable news |
+| distortion_rate | city | tendency of message to mutate |
+| market_sensitivity | city/good | how quickly prices react |
+| speech_weight | actor | how strongly others believe speaker |
+| secrecy_pressure | scenario | incentives to hide truth |
+
+---
+
+## 6. Speech Weight
+
+Not all voices carry equally.
+
+A claim from a respected man may outweigh three correct claims from porters.
+
+Speech weight affected by:
+
+- AVCTORITAS
+- FAMA
+- office held
+- wealth display
+- proven past accuracy
+- group prejudice
+- confidence of delivery
+
+Thus:
+
+A freedman may know first.  
+A noble may be believed first.
+
+---
+
+## 7. Information Delay
+
+Truth moves through time.
+
+Example: Capua timber fire.
+
+| Actor | Delay |
+|---|---|
+| contractor courier | 1–2 days |
+| connected merchant | 2–4 days |
+| ordinary market trader | 3–6 days |
+| distant rural buyer | longer |
+
+The actor with shorter delay can profit.
+
+---
+
+## 8. Distortion Mechanics
+
+As stories move, they mutate.
+
+```text
+distance ↑ -> distortion ↑
+retellings ↑ -> distortion ↑
+panic ↑ -> distortion ↑
+political_interest ↑ -> distortion ↑
+trusted_witness_present -> distortion ↓
+multiple_independent_reports -> distortion ↓
+```
+
+Example:
+
+Small warehouse fire becomes “entire district lost.”
+
+---
+
+## 9. Price Reaction Model
+
+```text
+price_change =
+scarcity_expectation
++ fear
++ hoarding
++ transport_uncertainty
+- trusted_reassurance
+- visible_replacement_supply
+```
+
+Meaning:
+
+Even false rumors can raise prices if believed long enough.
+
+---
+
+## 10. Merchant Use Cases
+
+### 10.1 Good Use
+
+Actor asks:
+
+- what is certainly known?
+- who benefits from this story?
+- what second-order shortage follows?
+- who has independent confirmation?
+- how long before truth spreads?
+
+### 10.2 Bad Use
+
+Actor asks only:
+
+- is it true?
+
+This is too narrow. Even false rumors can create real temporary opportunity.
+
+---
+
+## 11. Example: Bronze Forge Fire
+
+### Immediate Known Signal
+
+Smoke visible.
+
+### Rumors Within Hours
+
+- bronze forge destroyed
+- owner insolvent
+- sabotage by iron interests
+- workers dead
+- stock saved secretly
+- magistrate forcing sale
+
+### Rational Merchant Questions
+
+- tool supply reduced?
+- rebuild timber demand rising?
+- creditors exposed?
+- iron substitute demand imminent?
+- who knows the stock survived?
+
+---
+
+## 12. Example Dialogue Logic
+
+> “I heard the forge burned.”
+
+Low value statement.
+
+> “I saw smoke, three collapsed beams, and carts removing molds.”
+
+High value statement.
+
+> “I sold nails before noon.”
+
+Highest value statement — reveals belief through action.
+
+---
+
+## 13. Roman Land and Physical Detail in Rumor
+
+Rumor often uses concrete measures.
+
+Use Roman units:
+
+- IUGERUM (land area)
+- PASSUS (distance)
+- LIBRA (weight)
+- MODIVS (dry measure)
+
+Example:
+
+> “They mean to plant six iugera behind the forge for shaft wood.”
+
+This is stronger than vague speech.
+
+Specificity increases believability, even when false.
+
+---
+
+## 14. Social Filtering by Background
+
+| Background | Hears Best |
+|---|---|
+| Former Legionary | movement disruption, guard failures |
+| Freedman Trader | distress selling, salvage, street truth |
+| Noble Younger Son | family scandal, patron moves |
+| Failed Magistrate | debts, permits, seizures |
+| Camp Logistician | shortages, carts, labour demand |
+| Guild Scribe | insolvency, collateral, unpaid accounts |
+
+Same city, different information worlds.
+
+---
+
+## 15. Simulation Rules
+
+### Do Not Give Perfect Knowledge
+
+Actors should infer, not receive certainty.
+
+### Do Not Make Rumor Pure Randomness
+
+Rumor must emerge from actual events, incentives, and networks.
+
+### Do Not Make Truth Instantly Win
+
+Falsehood can dominate briefly.
+
+### Reward Verification
+
+Independent confirmation should create advantage.
+
+---
+
+## 16. Relations
+
+```text
+rumor_velocity ↑ -> price_adjustment_speed ↑
+distortion_rate ↑ -> false_opportunity ↑
+information_delay ↓ -> merchant_edge ↑
+speech_weight ↑ -> market_reaction ↑
+multiple_sources_confirm ↑ -> rumor_accuracy ↑
+panic ↑ -> hoarding ↑
+```
+
+---
+
+## 17. Repository Use
+
+Internal simulation substrate. Not player-facing prose.
+
+Use to support:
+
+- prologue BALNEA dialogue
+- city event reactions
+- price shocks
+- hidden information systems
+- actor asymmetry
+- scenario chaining
+- merchant skill differentiation
+
+---
+
+## 18. Canonical Success Condition
+
+If the participant stops asking:
+
+“Is the rumor true?”
+
+and starts asking:
+
+“Who heard it first, who profits if believed, and what changes before certainty arrives?”
+
+then this document is functioning correctly.