Symbolic Structurology as Symbolic Dynamics: A Three-Tier Classification and Convergence Analysis of 33 Cross-Cultural Symbol Pairs

Traditional semiotics treats the symbol as a meaning-bearing sign and describes cross-cultural difference qualitatively. This paper reports a computational study within the framework of Symbolic Structurology, a branch of Daopology that defines a symbol as a measurable structural unit (a kernel) and replaces meaning interpretation with structural comparison. We extend the static structural measure into a dynamic one by scoring each symbol along 20 fixed semantic dimensions at two time slices—a pre-modern "historical" layer and a present-day "current" layer—and by deriving a convergence ratio from the change in the structural difference measure D. Using a large language model (LLM) as the primary scorer and human review as the final arbiter, we analyze 33 Chinese–English (and ancient–modern) symbol pairs. Three results emerge. First, cross-cultural symbols fall into three well-separated tiers: undifferentiated (n = 7, Dhist < 0.09), semi-differentiated (n = 5, Dhist ∈ [0.23, 0.40]), and differentiated (n = 21, Dhist ∈ [0.32, 0.62]). Second, among the 26 differentiated or semi-differentiated pairs, convergence ratios concentrate between 0.322 and 0.844 (median 0.506), indicating a persistent, medium-intensity convergence rather than either complete fusion or complete stasis. Third, the LLM's directional judgments exhibit a systematic central-tendency bias: all 23 differentiated pairs were labelled "bidirectional convergence", even where historical evidence points to unidirectional borrowing. We therefore treat the directional field as diagnostic rather than conclusive and reserve it for human adjudication. The framework is positioned as a formal, falsifiable extension of semiotics, not a replacement for interpretive methods.