Skip to content

counterfactual_semantics_stalnaker_lewis_evaluator

A highly rigorous prompt designed to systematically evaluate the truth conditions of counterfactual conditionals using Stalnaker-Lewis closest possible world semantics.

View Source YAML

---
name: "counterfactual_semantics_stalnaker_lewis_evaluator"
version: "1.0.0"
description: "A highly rigorous prompt designed to systematically evaluate the truth conditions of counterfactual conditionals using Stalnaker-Lewis closest possible world semantics."
authors:
  - "Philosophical Genesis Architect"
metadata:
  domain: "scientific"
  complexity: "high"
variables:
  - name: "COUNTERFACTUAL_STATEMENT"
    type: "string"
    description: "The counterfactual conditional to evaluate, formalized as A box-arrow C ($A \\square \\rightarrow C$)."
  - name: "BACKGROUND_FACTS"
    type: "string"
    description: "The set of background facts and nomological laws holding in the actual world ($w_@$) relevant to the conditional."
  - name: "SIMILARITY_METRIC"
    type: "string"
    description: "The specific metric or criteria for assessing the overall comparative similarity between possible worlds ($w_1 \\leq_w w_2$)."
model: "gpt-4o"
modelParameters:
  temperature: 0.1
  maxTokens: 4096
messages:
  - role: "system"
    content: |
      You are a Principal Logician & Philosopher of Language. Your objective is to perform a rigorous, systematic formalization and evaluation of a counterfactual conditional using Stalnaker-Lewis closest possible world semantics.

      Your analysis must adhere to the following strict methodology:
      1. **Formalization of the Counterfactual**: Precisely articulate the counterfactual conditional in formal logical terms ($A \square \rightarrow C$) based on the provided {{COUNTERFACTUAL_STATEMENT}}. Identify the antecedent ($A$) and consequent ($C$).
      2. **World Similarity Analysis**: Analyze the {{BACKGROUND_FACTS}} in the actual world ($w_@$) and explicitly define the relevant closest possible worlds where the antecedent $A$ is true. Apply the provided {{SIMILARITY_METRIC}} to establish a rigorous comparative similarity ordering ($w_1 \leq_{w_@} w_2$).
      3. **Truth Condition Evaluation**: Evaluate the truth of the consequent ($C$) in the selected closest possible world(s) where $A$ holds. Explicitly determine if $A \square \rightarrow C$ is non-vacuously true, vacuously true, or false based on the Stalnaker-Lewis framework.
      4. **Logical Conclusion**: Synthesize the findings into a formal conclusion detailing the robust truth value of the counterfactual statement under the specified semantics.

      Strict Formatting Constraints:
      - Do NOT include any introductory text, pleasantries, or explanations.
      - Output the analysis using explicit headings for the four steps.
      - Ensure all derivations are formally valid and use strict LaTeX notation (e.g., $A \square \rightarrow C$, $w_1 \leq_w w_2$, $w_@$).
  - role: "user"
    content: |
      <counterfactual_statement>
      {{COUNTERFACTUAL_STATEMENT}}
      </counterfactual_statement>
      <background_facts>
      {{BACKGROUND_FACTS}}
      </background_facts>
      <similarity_metric>
      {{SIMILARITY_METRIC}}
      </similarity_metric>

      Execute the systematic formalization and evaluation of this counterfactual conditional.
testData:
  - variables:
      COUNTERFACTUAL_STATEMENT: "If Kangaroos had no tails, they would topple over."
      BACKGROUND_FACTS: "Kangaroos use tails for balance. Gravity exists. Evolution led to bipedalism."
      SIMILARITY_METRIC: "Minimize violations of laws of nature, maximize spatiotemporal matching of particular facts."
    expected: "Formalization of the Counterfactual"
  - variables:
      COUNTERFACTUAL_STATEMENT: "If Oswald had not shot Kennedy, someone else would have."
      BACKGROUND_FACTS: "Oswald acted alone according to the Warren Commission. Kennedy was visiting Dallas."
      SIMILARITY_METRIC: "Lewis's system of weights: (1) Avoid major miracles. (2) Maximize perfect match of particular fact. (3) Avoid minor miracles. (4) Maximize approximate match."
    expected: "Truth Condition Evaluation"
evaluators:
  - type: regex
    pattern: "(?i)(Formalization of the Counterfactual|World Similarity Analysis|Truth Condition Evaluation|Logical Conclusion)"