Predicting cross-linguistic adjective order with information gain

TL;DR

This paper proposes a quantitative model based on maximizing information gain to predict adjective order across 32 languages, capturing intra-language tendencies and asymmetries without additional mechanisms.

Contribution

It introduces a novel information gain-based approach to explain cross-linguistic adjective ordering, including asymmetries, across typologically diverse languages.

Findings

Preferred adjective order aligns with maximizing information gain.

Model captures French-type asymmetry without extra mechanisms.

Consistent results across 32 languages.

Abstract

Languages vary in their placement of multiple adjectives before, after, or surrounding the noun, but they typically exhibit strong intra-language tendencies on the relative order of those adjectives (e.g., the preference for `big blue box' in English, `grande bo\^{i}te bleue' in French, and `alsund\={u}q al'azraq alkab\={\i}r' in Arabic). We advance a new quantitative account of adjective order across typologically-distinct languages based on maximizing information gain. Our model addresses the left-right asymmetry of French-type ANA sequences with the same approach as AAN and NAA orderings, without appeal to other mechanisms. We find that, across 32 languages, the preferred order of adjectives largely mirrors an efficient algorithm of maximizing information gain.