Split verbal response strategy from voice style

docs/architecture/agent-state-distributions-and-prompt-projection.md

@@ -122,10 +122,7 @@ These describe how the performed self comes out in language:
 - lyricism
 - narrative detail
 - emotional explicitness
-- sarcastic deflection
 - pointedness
-- sincerity evasion
-- emotional evasion
 - self disclosure
 - hedging
 - certainty marking
@@ -136,21 +133,28 @@ These describe how the performed self comes out in language:
 - dialect marking
 - theatricality
 - humor
-- banter as boundary
-- verbal aggression
 - conversational dominance
 - listening responsiveness
 - question asking
 - profanity
 
-This layer matters because presentation strategy and voice are related but not
-identical. Performing a prickly self may come out as dry sarcasm, clipped
-literalism, theatrical disdain, or a flat refusal to offer emotional handles.
-If the schema only stores the performed self, the renderer has to guess the
-voice. That is how everyone becomes the same quippy little goblin with
-different hats.
+### 6. Verbal Response Strategy
 
-### 6. Situational State
+These describe pressure-activated speech moves:
+
+- sarcastic deflection
+- sincerity evasion
+- emotional evasion
+- banter as boundary
+- verbal aggression
+
+This layer matters because presentation strategy, voice, and verbal response
+are related but not identical. Performing a prickly self may come out as dry sarcasm, clipped
+literalism, theatrical disdain, or a flat refusal to offer emotional handles.
+If the schema only stores the performed self, the renderer has to infer voice
+and stress response from the same field.
+
+### 7. Situational State
 
 These are transient activations:
 
@@ -165,7 +169,7 @@ These are transient activations:
 - acute shame
 - perceived status threat
 
-### 7. Relationship State
+### 8. Relationship State
 
 Per target person, faction, group, product domain, or institution:
 
@@ -181,7 +185,7 @@ Per target person, faction, group, product domain, or institution:
 - expectation of care
 - expectation of betrayal
 
-### 8. Goals And Values
+### 9. Goals And Values
 
 - active goals
 - protected values

docs/architecture/agent-state-variable-glossary.md

@@ -165,10 +165,7 @@ quirks in the current examples.
 | `lyricism` | Uses musical, poetic, sensuous, or rhythmically heightened language. |
 | `narrative_detail` | Explains through context, sequence, anecdote, or concrete scene detail. |
 | `emotional_explicitness` | Names feelings, needs, wounds, affection, fear, or attachment directly. |
-| `sarcastic_deflection` | Uses sarcasm to redirect, dodge, or blunt sincerity, fear, gratitude, or pain. |
 | `pointedness` | Cuts directly, sharply, or with barbed precision. |
-| `sincerity_evasion` | Avoids emotionally plain or vulnerable language even when the feeling is present. |
-| `emotional_evasion` | Avoids naming, dwelling in, or accepting emotional baggage. |
 | `self_disclosure` | Volunteers personal history, inner state, motives, or private stakes. |
 | `hedging` | Softens claims with uncertainty markers, caveats, deference, or exits. |
 | `certainty_marking` | Signals confidence, finality, authority, or refusal to leave claims open. |
@@ -179,13 +176,25 @@ quirks in the current examples.
 | `dialect_marking` | Shows regional, class, occupational, subcultural, or community-specific speech markers. |
 | `theatricality` | Performs speech with heightened drama, staging, persona, flourish, or rhetorical display. |
 | `humor` | Uses jokes, wit, absurdity, teasing, or comic framing as a regular speech tool. |
-| `banter_as_boundary` | Uses joking exchange to maintain distance while staying socially engaged. |
-| `verbal_aggression` | Uses language to attack, intimidate, punish, or dominate. |
 | `conversational_dominance` | Takes, holds, redirects, or controls conversational floor and agenda. |
 | `listening_responsiveness` | Reflects, tracks, validates, or adapts to what the other person just said. |
 | `question_asking` | Uses questions to probe, invite, corner, teach, test, or keep the other person talking. |
 | `profanity` | Uses taboo, vulgar, sacred, or deliberately coarse language. |
 
+## Verbal Response Strategy
+
+Verbal response strategy describes pressure-activated speech moves. These are
+not baseline voice texture; they are ways speech manages exposure, threat,
+attachment pressure, shame, status, or control.
+
+| Label | Meaning |
+| --- | --- |
+| `sarcastic_deflection` | Uses sarcasm to redirect, dodge, or blunt sincerity, fear, gratitude, or pain. |
+| `sincerity_evasion` | Avoids emotionally plain or vulnerable language even when the feeling is present. |
+| `emotional_evasion` | Avoids naming, dwelling in, or accepting emotional baggage. |
+| `banter_as_boundary` | Uses joking exchange to maintain distance while staying socially engaged. |
+| `verbal_aggression` | Uses language to attack, intimidate, punish, or dominate. |
+
 ## Situational State
 
 | Label | Meaning |

docs/architecture/canonical-agent-state-schema.md

@@ -49,6 +49,7 @@ Each agent has:
 - `behavioral_dimensions`
 - `presentation_strategy`
 - `voice_style`
+- `verbal_response_strategy`
 - `situational_state`
 - `values`
 
@@ -95,7 +96,7 @@ The behavioral dimensions include the known must-not-drop set:
 
 Those are not garnish. They are load-bearing social handles.
 
-## Presentation Strategy Vs Voice Style
+## Presentation, Voice, And Response
 
 `presentation_strategy` describes the self being performed.
 
@@ -122,10 +123,7 @@ The required voice handles are:
 - `lyricism`
 - `narrative_detail`
 - `emotional_explicitness`
-- `sarcastic_deflection`
 - `pointedness`
-- `sincerity_evasion`
-- `emotional_evasion`
 - `self_disclosure`
 - `hedging`
 - `certainty_marking`
@@ -136,26 +134,31 @@ The required voice handles are:
 - `dialect_marking`
 - `theatricality`
 - `humor`
-- `banter_as_boundary`
-- `verbal_aggression`
 - `conversational_dominance`
 - `listening_responsiveness`
 - `question_asking`
 - `profanity`
 
+`verbal_response_strategy` describes pressure-activated speech moves:
+
+- `sarcastic_deflection`
+- `sincerity_evasion`
+- `emotional_evasion`
+- `banter_as_boundary`
+- `verbal_aggression`
+
 So Cat's prickliness is not only a dialogue prompt instruction. It is backed by
 canonical state:
 
 - high `presentation_strategy.abrasive_boundary`
 - high `presentation_strategy.ironic_distance`
-- high `voice_style.sarcastic_deflection`
-- high `voice_style.sincerity_evasion`
-- high `voice_style.emotional_evasion`
+- high `verbal_response_strategy.sarcastic_deflection`
+- high `verbal_response_strategy.sincerity_evasion`
+- high `verbal_response_strategy.emotional_evasion`
 
-The renderer can then translate that state into lines that dodge sincerity with
-pointed sarcasm without making her cruel at random. The goal is not "be snarky."
-The goal is "protect the soft tissue by making the other person deal with the
-spines first."
+The renderer can then combine the performed self, baseline voice, active stress,
+and response strategy into dialogue without treating every defensive move as a
+permanent voice trait.
 
 ## Perceived State
 

examples/agent-state.call-of-the-void.json

@@ -329,36 +329,11 @@
             "plasticity": 0.31,
             "current_activation": 0.78
           },
-          "sarcastic_deflection": {
-            "mean": 0.84,
-            "plasticity": 0.2,
-            "current_activation": 0.93
-          },
           "pointedness": {
             "mean": 0.77,
             "plasticity": 0.25,
             "current_activation": 0.82
           },
-          "sincerity_evasion": {
-            "mean": 0.8,
-            "plasticity": 0.24,
-            "current_activation": 0.87
-          },
-          "emotional_evasion": {
-            "mean": 0.78,
-            "plasticity": 0.27,
-            "current_activation": 0.85
-          },
-          "banter_as_boundary": {
-            "mean": 0.74,
-            "plasticity": 0.31,
-            "current_activation": 0.8
-          },
-          "verbal_aggression": {
-            "mean": 0.46,
-            "plasticity": 0.42,
-            "current_activation": 0.5
-          },
           "plainspoken_directness": {
             "mean": 0.62,
             "plasticity": 0.33,
@@ -469,6 +444,33 @@
             "plasticity": 0.38,
             "current_activation": 0.57
           }
+        },
+        "verbal_response_strategy": {
+          "sarcastic_deflection": {
+            "mean": 0.84,
+            "plasticity": 0.2,
+            "current_activation": 0.93
+          },
+          "sincerity_evasion": {
+            "mean": 0.8,
+            "plasticity": 0.24,
+            "current_activation": 0.87
+          },
+          "emotional_evasion": {
+            "mean": 0.78,
+            "plasticity": 0.27,
+            "current_activation": 0.85
+          },
+          "banter_as_boundary": {
+            "mean": 0.74,
+            "plasticity": 0.31,
+            "current_activation": 0.8
+          },
+          "verbal_aggression": {
+            "mean": 0.46,
+            "plasticity": 0.42,
+            "current_activation": 0.5
+          }
         }
       },
       "goals": [
@@ -818,36 +820,11 @@
             "plasticity": 0.48,
             "current_activation": 0.31
           },
-          "sarcastic_deflection": {
-            "mean": 0.28,
-            "plasticity": 0.52,
-            "current_activation": 0.24
-          },
           "pointedness": {
             "mean": 0.24,
             "plasticity": 0.5,
             "current_activation": 0.22
           },
-          "sincerity_evasion": {
-            "mean": 0.62,
-            "plasticity": 0.42,
-            "current_activation": 0.69
-          },
-          "emotional_evasion": {
-            "mean": 0.72,
-            "plasticity": 0.36,
-            "current_activation": 0.78
-          },
-          "banter_as_boundary": {
-            "mean": 0.45,
-            "plasticity": 0.48,
-            "current_activation": 0.49
-          },
-          "verbal_aggression": {
-            "mean": 0.09,
-            "plasticity": 0.52,
-            "current_activation": 0.05
-          },
           "plainspoken_directness": {
             "mean": 0.4,
             "plasticity": 0.49,
@@ -958,6 +935,33 @@
             "plasticity": 0.39,
             "current_activation": 0.03
           }
+        },
+        "verbal_response_strategy": {
+          "sarcastic_deflection": {
+            "mean": 0.28,
+            "plasticity": 0.52,
+            "current_activation": 0.24
+          },
+          "sincerity_evasion": {
+            "mean": 0.62,
+            "plasticity": 0.42,
+            "current_activation": 0.69
+          },
+          "emotional_evasion": {
+            "mean": 0.72,
+            "plasticity": 0.36,
+            "current_activation": 0.78
+          },
+          "banter_as_boundary": {
+            "mean": 0.45,
+            "plasticity": 0.48,
+            "current_activation": 0.49
+          },
+          "verbal_aggression": {
+            "mean": 0.09,
+            "plasticity": 0.52,
+            "current_activation": 0.05
+          }
         }
       },
       "goals": [

schemas/agent-state.required-fields.json

@@ -89,10 +89,7 @@
       "lyricism",
       "narrative_detail",
       "emotional_explicitness",
-      "sarcastic_deflection",
       "pointedness",
-      "sincerity_evasion",
-      "emotional_evasion",
       "self_disclosure",
       "hedging",
       "certainty_marking",
@@ -103,12 +100,17 @@
       "dialect_marking",
       "theatricality",
       "humor",
-      "banter_as_boundary",
-      "verbal_aggression",
       "conversational_dominance",
       "listening_responsiveness",
       "question_asking",
       "profanity"
+    ],
+    "verbal_response_strategy": [
+      "sarcastic_deflection",
+      "sincerity_evasion",
+      "emotional_evasion",
+      "banter_as_boundary",
+      "verbal_aggression"
     ]
   }
 }

schemas/agent-state.schema.json

@@ -167,6 +167,7 @@
         "behavioral_dimensions",
         "presentation_strategy",
         "voice_style",
+        "verbal_response_strategy",
         "situational_state",
         "values"
       ],
@@ -195,6 +196,9 @@
         },
         "voice_style": {
           "$ref": "#/$defs/variable_map"
+        },
+        "verbal_response_strategy": {
+          "$ref": "#/$defs/variable_map"
         }
       }
     },

state/evidence.jsonl

@@ -12,3 +12,4 @@
 {"ts": "2026-05-01T18:26:40+00:00", "type": "design", "status": "accepted", "note": "Reworked perceived dimensions so observer reads do not know canonical mean or plasticity. Perceived state now tracks observed_activation, attributed_disposition, and confidence to model fundamental attribution bias; confidence is driven by evidence, emotion, stakes, memory, prejudice, and social reinforcement."}
 {"ts": "2026-05-01T18:35:51+00:00", "type": "design", "status": "accepted", "note": "Expanded voice_style from a narrow quirk set into a broad required communication vector covering register, pace, precision, disclosure, politeness, humor, dominance, responsiveness, and related speech mechanics."}
 {"ts": "2026-05-01T18:42:59+00:00", "type": "design", "status": "accepted", "note": "Added explicit voice_style axes for lyricism, coded_politeness, theatricality, and technical_compression so the required vector preserves the intended speech examples rather than folding them into weaker adjacent labels."}
+{"ts": "2026-05-01T18:49:11+00:00", "type": "design", "status": "accepted", "note": "Separated verbal_response_strategy from voice_style. Sarcastic deflection, sincerity evasion, emotional evasion, banter-as-boundary, and verbal aggression are pressure-activated speech defenses, not baseline voice texture."}

state/map.yaml

@@ -77,6 +77,7 @@ accepted_design:
   - Canonical authored character variables do not carry certainty; perceived dimensions do not expose canonical mean or plasticity, and instead model observed activation, attributed disposition, and confidence.
   - Fundamental attribution bias is first-class in perceived state: observers can mistake exceptional behavior for stable character, with stubborn reads emerging from confidence drivers such as memory, stakes, emotion, prejudice, and social reinforcement.
   - Voice style is a broad required communication vector covering lyrical speech, formality, verbosity, hedging, coded politeness, theatricality, dialect, register switching, technical compression, ritualized address, and related language mechanics.
+  - Verbal response strategy is separate from voice style; pressure-activated moves like sarcastic deflection, sincerity evasion, emotional evasion, banter-as-boundary, and verbal aggression model speech defenses rather than baseline voice texture.
 
 current_status:
   summary: Ghostlight now has the Epiphany-style persistence spine, the first architecture payload, and the initial canonical agent-state schema seam for Call of the Void-style authoring examples. The schema is still v0 and should be hardened before building prompt projection on top of it.

tools/validate_agent_state.py

@@ -101,6 +101,7 @@ def validate_agent(agent: dict[str, Any], required: dict[str, list[str]]) -> Non
             "behavioral_dimensions",
             "presentation_strategy",
             "voice_style",
+            "verbal_response_strategy",
             "situational_state",
             "values",
         ],
@@ -112,6 +113,7 @@ def validate_agent(agent: dict[str, Any], required: dict[str, list[str]]) -> Non
         "behavioral_dimensions",
         "presentation_strategy",
         "voice_style",
+        "verbal_response_strategy",
         "situational_state",
     ]:
         validate_variable_map(canonical[group], required[group], f"{path}.canonical_state.{group}")