Search for a command to run...
This pre-print presents a foundational, representation-agnostic framework for semantic interpretation grounded in topology and compositional mapping. Rather than modeling meaning within a single representational space, the paper introduces the notion of a meaning topology as a general semantic domain in which semantic connectivity arises through two structurally distinct modes: causal meaning, defined by direct participation in generative subgraphs, and reflective meaning, defined by interpretative mappings across the topology. The framework formally demonstrates that causal and reflective meanings are non-equivalent, may coexist, and cannot be exhaustively captured by a single reflection operator. Metric embedding spaces commonly used in distributional semantics are shown to constitute restricted special cases of the proposed theory, thereby clarifying both their empirical utility and their structural limitations. A central contribution of the paper is the formalization of hierarchy of meaning as compositional mapping between distinct meaning topologies, rather than as internal stratification within a single space. This formulation imposes no assumptions of inclusion, complexity ordering, or linear abstraction, and supports branching, collapsing, and recursive hierarchical constructions. By shifting the focus from representation to interpretation, and from single-space semantics to multi-topology compositional structure, the paper provides an axiomatic foundation for semantic plurality and hierarchical meaning across linguistic, cognitive, and artificial systems. The work is intended as a theoretical contribution to the foundations of semantics, artificial intelligence, and philosophy of information, and does not rely on empirical datasets or experimental evaluation. Keywords:semantic topology, causal meaning, reflective meaning, hierarchy of meaning, semantic foundations, interpretative mapping, artificial intelligence