Subject-Relative Cognitive Languages and Translation Dynamics--Reconstructing Explainability, Mathematical Nature, Objectivity, and Co-Being

Communication between different cognitive agents—human or AI—can be modeled as partial translation rather than perfect understanding. Each agent has its own classification structure and cognitive language. The paper formalizes preservation obligations, translation incompleteness, and local objectivity, and proposes reconstructing the definitions of cognitive identity, qualia, and objectivity from the standpoint of translatability. BrainOS and its plugins are presented as the practical foundation for this theory.

Study at a glance

Abstract

This paper proposes a theoretical framework for understanding communication between different cognitive agents, including humans and AI, by assuming that each agent possesses its own classification structure and cognitive language. Rather than treating understanding as perfect identity, it models it as partial translation. The paper formalizes preservation obligations, translation incompleteness, and local objectivity, and proposes reconstructing the definitions or axioms of cognitive identity, qualia, and objectivity from the standpoint of translatability. BrainOS and its various plugins are presented as the practical foundation of this theory. 本稿は、人間とAIを含む異なる認知主体が、それぞれ固有の類別構造と認知言語を持つことを前提に、主体間の理解を「完全一致」ではなく部分翻訳として捉える理論を提案する。保存責務、翻訳不完全性、局所的客観性を形式化し、認知同一性・クオリア・客観性の定義や公理を、翻訳可能性から再構成する研究方向を示す。BrainOSと各種Pluginの実装事例も、この理論の実践的基盤として位置づける。