Non-Logical Transitions as the Minimal Source of Information, Time, and Hierarchical Irreversibility
Author: Jia Baolong
Date: March 2026
Status: Conceptual/Theoretical Draft
Abstract
We propose a minimal generative principle for evolving systems: any system capable of sustained information production must contain self-referential structure and, consequently, points of undecidability. We show that continued evolution across such points requires non-logical state transitions—updates not derivable from the system’s prior state description. These transitions introduce irreducible information and induce a natural partial ordering interpreted as time. We further propose a hierarchical irreversibility law: once self-reference is lost at a given layer, it cannot be reconstructed by higher layers. The framework does not assume physical laws and is intended as a candidate foundation for generative systems rather than a complete physical theory.
1. Introduction
A central question across physics, computation, and complex systems is the origin of genuinely new information. Deterministic systems preserve information under complete description, while stochastic models typically assume randomness rather than derive it.
We investigate the minimal structural requirements for a system to generate information endogenously, without introducing external randomness or predefined physical laws.
2. Generative Constraint
Proposition 1 (Generative Constraint).
A system capable of unbounded information production must contain a self-referential description that is not logically complete.
Rationale.
A fully closed and complete system is compressible and cannot generate irreducible novelty. Self-reference introduces the possibility of undecidable propositions (Gödel-type structures), preventing closure.
3. Minimal Structural Triad
We define three necessary components of any generative system:
- SR (Self-Reference): The system can encode and evaluate statements about its own state.
- ER (Entity-Relation): A distinguishable state space consisting of entities and relations.
- NC (Non-Closure): Points where evaluation cannot be completed within the system’s inference rules.
We emphasize that NC replaces informal notions such as “lazy evaluation” and is defined as a structural property, not an implementation detail.
4. Non-Logical Transitions
Theorem 1 (Generative Transition Necessity).
If a self-referential system continues evolving across points of non-closure, it must perform state transitions not derivable from its prior state description.
Interpretation.
At undecidable points, neither continuation nor resolution is logically determined. Any continued evolution therefore introduces a transition that cannot be reduced to prior rules.
5. Information Emergence
Theorem 2 (Irreducible Information Injection).
Each non-logical transition introduces information not contained in the system’s prior description.
This follows from the fact that the transition outcome is not computable from the prior state.
6. Emergent Time
Definition (Generative Time).
Time is the partial order induced by successive non-logical transitions.
Properties: - Irreversibility: prior transitions cannot be undone without loss of introduced information - Directionality: ordering is defined by dependency of states
7. Hierarchical Structure
We consider systems composed of layered abstractions:
- Base layer: contains SR and enables generative transitions
- Derived layers: may perform computations without full self-reference
8. Irreversibility of Self-Reference
Proposition 2 (SR Non-Reconstruction Law).
If a layer loses the capacity for self-reference, no higher layer built upon it can reconstruct true self-reference.
Interpretation.
Self-reference is not an emergent property that can be reintroduced from purely derived computation; it must be preserved through the generative base.
9. ER₀ Structures
Definition (ER₀).
A structure consisting purely of entity-relations without self-reference.
Properties: - No intrinsic time - No endogenous information generation - Can be represented but not internally evolved
10. Discussion
This framework suggests that:
- Information generation requires structural incompleteness
- Time arises from non-logical transitions rather than predefined dynamics
- Hierarchical systems inherit constraints from their generative base
We do not claim equivalence with physical reality but propose a minimal architecture for systems exhibiting open-ended complexity.
11. Limitations
- The framework is conceptual and not yet formalized as a complete mathematical system
- No empirical validation is provided
- Relationship to physical theories remains speculative
12. Conclusion
We propose that non-logical transitions induced by self-referential incompleteness constitute the minimal mechanism for generating information, time, and hierarchical irreversibility. This provides a candidate foundation for studying generative systems across disciplines.
References
Gödel, K. (1931). On formally undecidable propositions.
Turing, A. (1936). On computable numbers.
Shannon, C. (1948). A mathematical theory of communication.