From language development to language evolution: A unified view of human lexical creativity

Area of Science:

• Cognitive Science
• Linguistics
• Computational Linguistics

Background:

• Human language uniquely employs word meaning extension for creative expression.
• Lexical creativity appears across different timescales, from individual development to language evolution.

Purpose of the Study:

• To investigate if diverse manifestations of lexical creativity share a common underlying foundation.
• To determine if computational models can capture commonalities between developmental and evolutionary semantic data.

Main Methods:

• Utilized computational modeling to analyze semantic knowledge.
• Applied models to both child language development data and historical word meaning evolution data.
• Examined data spanning over 1400 languages for evolutionary insights.

Main Results:

• A parsimonious set of semantic knowledge types effectively characterized both developmental and evolutionary data.
• Models trained on evolutionary data accurately predicted developmental patterns, and vice versa.
• Demonstrated a strong cross-predictive capability between developmental and evolutionary semantic models.

Conclusions:

• Suggests a unified foundation for human lexical creativity.
• Highlights common semantic principles underlying both individual language acquisition and long-term language change.
• Provides evidence for shared mechanisms driving meaning extension across ontogeny and phylogeny.