TLDR: This research paper introduces a formal framework to unify the ambiguous concept of “state” across various domains. It proposes a “hierarchical state grid” to represent all states (mathematical, physical, linguistic) and an “Intermediate Meta-Universe” (IMU) to manage meta-level operations like language translation, agent integration, and real-time changes, avoiding self-referential paradoxes. The framework also distinguishes between “proof” and “verification” through “inter-universal operations,” emphasizing that all definitions must be verifiable. This work aims to provide a robust foundation for defining intelligence and advancing formal logic and scientific theory.
The concept of “state” is fundamental across many scientific and mathematical disciplines, yet it has long suffered from a lack of a unified and precise definition. This ambiguity poses a significant challenge, particularly for rigorous theoretical constructions like the axiomatic definition of intelligence. A recent research paper by Kei Itoh addresses this critical issue by proposing a novel formal framework that aims to bring clarity and consistency to the concept of state.
A Unified View of “State” with the Hierarchical State Grid
At the core of this new framework is the “hierarchical state grid,” a two-dimensional coordinate system designed to represent any type of state uniformly. This grid features two key axes: “state depth” and “mapping hierarchy.” State depth refers to the level of complexity, where more complex states are defined based on simpler, lower-order ones. Mapping hierarchy, on the other hand, indicates the order of a conceptual mapping, ranging from 0th-order states (simple labels) to higher-order mappings (mappings whose domains are themselves mappings).
This innovative grid allows for the encoding of all states—whether from physics, logic, or language—onto a single coordinate system. A significant outcome of this approach is the unification of several previously distinct concepts: “definition,” “state,” and “structure.” The paper demonstrates that the act of defining something can be viewed as a specific type of state, a “definitional mapping.” Similarly, what we traditionally call a “structure” is reinterpreted as a “predicate state” within this grid, eliminating the need for a separate definition of structure. This means that determining a state and determining a definition for an object are essentially the same operation.
The grid also introduces “definability” as a fundamental state at its lowest layer (depth 0). This seemingly simple concept becomes crucial when considering “real time,” as it allows the framework to systematically accommodate unobserved, unknowable, or future phenomena that are not yet definable in the present moment.
The Intermediate Meta-Universe: Bridging Theory and Reality
To make this formal framework robust and avoid logical inconsistencies that arise from self-reference (like Gödel’s incompleteness theorems), the paper introduces the “Intermediate Meta-Universe” (IMU). The IMU acts as a crucial buffer layer between the “true meta-universe” (where we, as definers, exist) and the “definition universe” (where our definitions reside). By creating “mirror-like objects” from entities in the true meta-universe, the IMU allows meta-level concepts—such as definers themselves, the languages they use, and even the definition of definitions—to be treated as explicit objects within the theory, without triggering paradoxes.
The IMU serves as a powerful “external operation API” for the definition universe, enabling three vital functions:
- Translation Mapping: It manages the coexistence of different definition languages (e.g., set theory, category theory, natural language, programming languages). The IMU formalizes how concepts and states can be translated and transferred between these diverse linguistic systems, ensuring that the essence of a definition is preserved across different representations.
- Integration Mapping: In a world where multiple agents (humans or AI) observe and edit definitions, conflicts are inevitable. The IMU provides a mechanism to safely and consistently integrate edits and differences from various agents, similar to version control systems. It allows for clear visualization of aligned and misaligned states, facilitating knowledge integration, especially for meta-level communication between AI systems.
- Real-Time Mapping: The definition universe is not static; it evolves with the progression of “real time.” The IMU introduces real-time mapping to explicitly capture these changes. This function also clarifies the distinction between “virtual time” (a numerically structured, definable time within a definition) and “real time” (a continuous, meta-level concept that cannot be fully defined). This distinction is key to understanding the difference between “proof” (static validation) and “verification” (dynamic validation) in scientific theories. The 3D extension of the hierarchical state grid, incorporating time as a dimension, further enhances this capability, allowing for pseudo-descriptions of future states and the management of unobserved information.
Also Read:
- Navigating the Mathematical Landscape: LLMs in Formal and Informal Reasoning
- Generative Logic: Automating Mathematical Discovery Through a New Computing Paradigm
Inter-Universal Operations: Connecting Diverse Systems
Building on the meta-universe theory, the paper elaborates on “inter-universal operations,” which are transformations across different universes—be they axiomatic systems, languages, agents, or time. These operations are categorized into two main types:
- Macrocosm-inter-universal operations: These involve global transformations that span entire language systems or axiomatic frameworks, preserving or transforming universal structures. They are less affected by real-time progression and are typically associated with mathematical “proof.”
- Microcosm-inter-universal operations: These are localized transformations that map specific concepts or structures between different theoretical universes. They are inherently influenced by real-time progression and are central to the process of “verification” in non-mathematical fields like science and engineering.
A crucial aspect of these operations is the “codomain requirement”: all definitions, regardless of their initial form, must ultimately be transported into a universe where they can be formally proven, shared, or verified in real-time. This means ambiguous natural language proofs should be converted into machine-verifiable formal languages, and naive concepts should be transformed into forms suitable for mechanical implementation, such as programming languages, to validate their practical utility.
In essence, this research presents a meta-formal logical framework, grounded in the principle that “definition equals state,” that spans time, language, agents, and operations. It provides a mathematically robust foundation applicable to the definition of intelligence, formal logic, and scientific theory at large. For more details, you can refer to the full research paper here.
