What question did this study set out to answer?

This research aims to develop a framework to understand how cultural meanings change over time, especially focusing on lexical markers.

May 18, 2026Open Access

Structural Hermeneutic Drift: A Quantitative Framework for Measuring Interpretive Frame Replacement in Cultural Transmission

Key Points

This research aims to develop a framework to understand how cultural meanings change over time, especially focusing on lexical markers.
Introduced Structural Hermeneutic Drift (SHD) as a framework for analyzing interpretive architecture changes.
Evaluated the framework using eleven historical case studies, assessing its internal consistency and coherence.
Compared SHD metrics with traditional embedding methods to highlight differences in analysis.
Achieved perfect ordinal separation between positive (7) and negative (3) cases, Mann-Whitney U = 21/21, exact p = 0.008.
Demonstrated that existing embedding methods risk conflating different types of structural changes.
Illustrated persistent spatial overlap in PCA and UMAP projections, confirming SHD’s unique interpretive capabilities.

Abstract

We introduce the Structural Hermeneutic Drift (SHD) framework, a computablemodel for decomposing and measuring the process by which the interpretive architecturesurrounding a culturally stable lexical marker is structurally replaced. Unlike diachronicword-embedding methods, which measure distributional distance between temporal snapshots,SHD takes the drift event as its unit of analysis: a phased trajectory from original framethrough friction, amplification, stabilization, and condensation into a replacement frame.The framework derives a suite of descriptive metrics—including Institutional AuthorityConcentration (ICA), Frame Compression Depth (FCD), and Drift Success Index (DSI)—from expert-coded historical inputs.We evaluate the framework’s internal consistency against eleven canonical historical cases(seven positive, three negative, one borderline), including two timestamped pre-coding heldoutevaluations (Judas Iscariot, Methuselah). The coding logic produces perfect ordinalseparation between POSITIVE and NEGATIVE cases (Mann-Whitney U = 21/21, exactp = 0.008), demonstrating strong internal coherence and establishing a foundation for largerscalevalidation. A general-purpose sentence-embedding baseline applied to the same textsillustrates the risk of conflating structural frame replacement with lexical variation. Acomparative analysis of PCA, t-SNE, and UMAP projections of OpenAI embeddings showspersistent spatial overlap between cases of structural drift (POSITIVE) and cases without drift(NEGATIVE), illustrating that latent geometry alone cannot recover the historical structureof frame replacement. Ultimately, SHD does not replace embedding-based semantic analysis;it constrains and contextualizes it through historical structural decomposition, providing aformal, auditable vocabulary for the comparative analysis of interpretive transformation.

Read Full Paperexternally

Mark Helpful

Bookmark

Relay

View Full Paper