On the Diachronic Stability of Irregularity in Inflectional Morphology

TL;DR

This paper investigates how irregular forms in language inflectional systems persist over time by simulating language learning with neural networks, focusing on the relationship between word frequency and irregularity.

Contribution

It introduces a neural network-based simulation approach to study the diachronic stability of irregular inflections in languages.

Findings

Irregular forms tend to persist when they are frequent.

Frequency influences the likelihood of irregularity surviving over time.

Neural simulations reveal conditions for the stability of irregular forms.

Abstract

Many languages' inflectional morphological systems are replete with irregulars, i.e., words that do not seem to follow standard inflectional rules. In this work, we quantitatively investigate the conditions under which irregulars can survive in a language over the course of time. Using recurrent neural networks to simulate language learners, we test the diachronic relation between frequency of words and their irregularity.