Status

Accepted

Context

The article "The Linguistic Sign: Metonymy and Virtuality" (DOI: 10.13092/lo.80.3565) by Radden and Kövecses provides a formal model for how linguistic signs activate meaning.

AME (Affinity Mapping Engine) from the Mythogen framework operates on a structurally identical mechanism. A single detected value activates the entire relational field. The overlap warrants documentation as an ADR.

Decision

Integrate the linguistic metonymy model as theoretical grounding for AME architecture. The trapezium model of virtual activation maps directly to AME Living Seed activation. The FOT (Field of Trust) operates as a virtual field, not a direct measurement.

Options Considered

Option	Description	Verdict
A	Full Radden/Kövecses trapezium integration	✅ Adopt — provides theoretical foundation for AME
B	Partial adoption (virtuality only)	⚠️ Insufficient — metonymic activation is the key mechanism
C	No integration	❌ Rejected — misses structural isomorphism with AME design

Positive

• AME single-element detection is architecturally sound.

Detecting one value should activate the whole seed. Because the sign works metonymically.

• FOT is a field (not a score).

Trust is virtually activated. Not directly measured.

• Trapezium structure provides a design pattern for expanding AME.

Any new element should virtually activate the full trapezium: form → concept → projections → relationships.

• The article distinction between real-world activation and virtual activation is the theoretical foundation for AME anti-capture design.

Negative

• Overfitting AME to the article model could lead to ignoring non-linguistic modalities (embodied, spatial, relational).
• V-crystal positions are not yet formally mapped to trapezium components.

Needs design work.

• The 30-day time lock may be too conservative or too lenient depending on community velocity.

Empirical tuning needed.

Risks

• V-crystal positions are not yet formally mapped to trapezium components.
• The 30-day time lock may be too conservative or too lenient depending on community velocity.

Article: Radden and Kövecses

DOI: https://doi.org/10.13092/lo.80.3565

Core claim: the linguistic sign means not by direct reference to the real world. It means by virtually activating a trapezium-like configuration of forms, concepts, experienced projections, and relationships. The process is metonymic (a part activates the whole) and virtual (indirect, not literal).

Form → Concept → Experienced Projections
         ↕
   Relationships (between all above)

The linguistic form (word) activates the concept. The concept in turn virtually activates experienced projections and the relational network. The whole configuration is engaged. Not just the literal referent.

Key insight: "activation of the real world remains dubious or indirect"

AME: Living Seed Activation Model

Single detected VALUE
  ↓ metonymically activates
Entire Living Seed (needs + beliefs + principles + values)
  ↓ virtually activates
FOT (virtual trust field)
  ↓ virtually activates
Affinity connections + matching

Direct Mapping

Article Concept	AME Equivalent
Linguistic form	Detected seed element (a value, a need, a belief)
Concept	The Living Seed — the whole person
Experienced projections	V-crystal position, maturity, bloom cycles
Relationships	Affinity links, trust radius, FOT indicators
Virtuality	FOT = virtual field, not directly observable
Metonymy	One detected element activates the whole seed
Trapezium	Living Seed structure: needs + beliefs + principles + values + state + FOT

Living Seed

A Living Seed is NOT a static user profile. It is a growing entity.

LivingSeed {
  needs: Vec        // survival, belonging, growth
  beliefs: Vec     // theories about reality
  principles: Vec // action guides
  values: Vec      // what is lived with others
  state: SeedState        // GROWING | DORMANT | BLOOMING | SEEDING
  maturity: u8            // 0-100
  fot: FOTRecord          // 5 indicators, virtual trust field
  v_crystal: VPosition    // Victor|Villain|Victim|Vengeful|Virtuous|Vulnerable
  affinities: Vec
  trust_radius: u8
}

The Four Distinctions

AME stores four distinct types of content. Never conflated.

Type	Changes	Stored Separately
Needs	Weekly	✅
Beliefs	Monthly	✅
Principles	Quarterly	✅
Values	Annually or slower	✅

Values require others to practice. The Desert Island Test: you cannot practice generosity alone. This is the relational foundation.

Trust Formula

Authentic Expression + Witnessed Resonance + Emotional Density = Trust Field (FOT)

FOT = Field of Trust. A virtual field. Not directly observable. Inferred from 5 indicators.

Indicator	Measures
Mutual Support	Unprompted help events
Response Velocity	Rally speed from request to response
Difficult Topic	Conflict raised and resolution rate
Benefit Distribution	Value flow Gini coefficient
Psychological Safety	Unsafe disclosures and safe responses

Hologram Principle: FOT composite = Math.min() of all 5 indicators. One off. The whole field degrades. Not averaged.

V-Crystal and the Vulnerability Axis

The 6 positions:

• Victor — wins/succeeds
• Villain — causes harm
• Victim — feels powerless
• Vengeful — seeks retaliation
• Virtuous — judges morally
• Vulnerable — circuit breaker (AXIS)

Key distinction (Vic Desotelle):

> "Vulnerability is the AXIS (not a point). > Virtuous and Vengeful are the POLES. > Victor, Villain, Victim spin around the axis. > Dynamic and alive."

Time Lock and Maturation

The 30-day time lock:

• Patterns detected less than 30 days ago are NOT used for matching.
• Seed maturity affects how much weight new signals carry.
• Trust cannot be gamed in short windows.

It accumulates through sustained virtual activation.

Open Questions

• Does the trapezium model suggest AME should also track "experienced projections" separately from values and beliefs?
• Can the cognitive grammar approach (Langacker) inform how AME generates Why-Cards?

Making explanations themselves metonymic?

• Should AME "membrane" (semi-permeable boundary) be mapped to the trapezium relationship layer?

References

• Article: "The Linguistic Sign: Metonymy and Virtuality" — DOI 10.13092/lo.80.3565, Radden and Kövecses, Linguistik Online, 2017
• AME Implementation Guide v0.2 (github.com/regentribes/mythogen-ame/docs/AME_IMPLEMENTATION_GUIDE.md)
• AME Architecture Mythogen (~/Projects/mythogen-ame/docs/AME-Architecture-Mythogen.md)