Grammatical Gender Disambiguates Syntactically Similar Nouns

Area of Science:

• Linguistics
• Cognitive Science
• Computational Linguistics

Background:

• Grammatical gender systems, though seemingly arbitrary, aid lexical prediction by reducing candidate word activation.
• Information theory perspectives suggest gender marking simplifies language processing.

Purpose of the Study:

• To investigate the role of syntax in shaping grammatical gender assignment cross-linguistically.
• To test the hypothesis that syntactically similar words are less likely to share gender.

Main Methods:

• Defined word syntactic distribution using probability vectors of dependency relations.
• Extracted syntactic relations for 32 languages from Universal Dependencies Treebanks.
• Employed correlational and mixed-effects regression analyses.

Main Results:

• Found that syntactically similar nouns are less likely to share grammatical gender across languages.
• Observed the opposite pattern for semantically and orthographically similar words.
• This contrasts with findings for semantic and orthographic similarity.

Conclusions:

• Grammatical gender assignment appears to be influenced by syntactic properties, not just semantic or orthographic ones.
• This suggests a functional design of language, optimizing for learnability and processing.
• The study contributes to understanding how cognitive pressures shape linguistic structures.