Earthling Logic – A Benchmark-Based Perspective of Truth

Reasoning under uncertainty, vagueness, and incomplete information remains a central challenge across artificial intelligence, philosophy, and decision science. While classical logic guarantees deductive rigor through binary abstractions, probabilistic models and fuzzy systems handle randomness and vagueness in isolation—frequently compromising classical deductive laws or leaving truth values ungrounded.This paper presents Earthling Logic (EL), a formal framework that establishes a benchmark-based perspective of truth to unify deterministic, vague, and stochastic reasoning within a single semantic architecture. By replacing abstract truth tables and ungrounded model-theoretic assignments with explicit operational reference points—a truth benchmark and a falsehood benchmark—EL grounds logical connectives and statemental operations in empirical credibility. This benchmark-driven semanticsenables continuous, graded, and probabilistic assessments while preserving the full deductive power of classical logic.Through Statemental Credibility Logic (SCL) and Predicate Credibility Logic (PCL), EL provides an axiomatic foundation that connects epistemological theories of truth with computational requirements. The framework offers a transparent, interpretable substrate for hybrid AI architectures, neurosymbolic integration, and automated decision-making under heterogeneous evidence.