A Bayesian Model for Discovering Typological Implications

TL;DR

This paper introduces a Bayesian computational model that automates the discovery of linguistic typological implications, including both known and novel insights, while addressing sampling issues inherent in language data.

Contribution

The paper presents a hierarchical Bayesian model that automates the discovery of typological implications, improving over manual analysis and handling language dependence.

Findings

Successfully identified well-known typological implications

Discovered novel implications for further linguistic study

Addressed the sampling problem in linguistic data analysis

Abstract

A standard form of analysis for linguistic typology is the universal implication. These implications state facts about the range of extant languages, such as ``if objects come after verbs, then adjectives come after nouns.'' Such implications are typically discovered by painstaking hand analysis over a small sample of languages. We propose a computational model for assisting at this process. Our model is able to discover both well-known implications as well as some novel implications that deserve further study. Moreover, through a careful application of hierarchical analysis, we are able to cope with the well-known sampling problem: languages are not independent.