# # #

Shane Steinert-Threlkeld & Jakub Szymanik

Presenters

Shane Steinert-Threlkeld & Jakub Szymanik

Title

Ease of learning explains semantic universals

Abstract

Despite extraordinary differences between natural languages, linguists have identified many semantic universals – shared properties of meaning – that are yet to receive a unified explanation. We analyze universals in a domain of content words (color terms) and a domain of function words (quantifiers). Using tools from machine learning, we show that meanings satisfying attested universals are easier to learn than those that are not. Thus, ease of learning can explain the presence of semantic universals in many different linguistic domains.

References