ChronoLens: Measuring Language Change Across Time, Languages, and Linguistic Levels

2026-08-04

AI Summary

This study addresses the challenge of systematically comparing historical language evolution across morphological, syntactic, semantic, and pragmatic levels within a unified framework. To this end, it proposes a novel analytical approach integrating frozen multilingual models, feature-aligned cross-encoders, and post-hoc linguistic interventions, enabling—for the first time—the quantification of multilingual, multi-period, and multi-level linguistic change in a shared representational space. Evaluated on a large-scale corpus comprising 44.98 million documents (approximately 17.2 billion tokens), the method yields sparse representations that correlate significantly more strongly with established linguistic metrics than dense embeddings or sparse autoencoders (ρ = 0.72 vs. 0.29/0.28). The results reveal that while the magnitude of internal change is comparable across languages, their evolutionary trajectories differ markedly, with substantial cross-linguistic variation in the timing, extent, and direction of change.

Abstract

Historical language change affects morphology, syntax, semantics, and pragmatics, yet computational studies typically examine these levels with incompatible representations and therefore cannot determine whether they evolve together across languages. We address this problem by asking how the magnitude and direction of change vary across linguistic levels, languages, and historical periods within a single analytical space. We introduce ChronoLens, a framework that combines frozen multilingual language models, feature-aligned crosscoders, and post-hoc linguistic interventions, and apply it to 44.98 million documents and approximately 17.2 billion tokens from five parliamentary traditions spanning 1803--2026. The resulting sparse representations agree substantially more strongly with linguistic statistics than dense embeddings or a pooled sparse autoencoder ($ρ=0.72$ versus $0.29$ and $0.28$), and reveal that morphology, syntax, semantics, and pragmatics generally change by comparable amounts within a language, while languages differ markedly in when, how far, and in which direction they change. These findings show that historical language change is a structured, multidimensional process: similar magnitudes can conceal different trajectories, and meaningful cross-linguistic comparison requires measuring both distance and direction.

Problem

• language change
• linguistic levels
• cross-linguistic comparison
• historical linguistics
• computational linguistics

Innovation

• ChronoLens
• language change
• multilingual modeling
• sparse representations
• cross-linguistic comparison