Beyond lexicostatistics: How to get more out of ‘word list’ comparisons

Abstract

This article surveys various long-standing ambiguities and confusions that continue to dog lexicostatistics and glottochronology. I aim to offer some novel perspectives and clarifications, which also help map out how we might devise new, alternative methods to build upon the good in Swadesh’s troubled legacy. I challenge the recent trend towards honing down Swadesh’s original list to a minimal core. A richer signal on language relationships is to be had not by discarding the data in meanings considered ‘unstable’, but by exploring the revealing patterns that emerge only when those meanings are kept, and contrasted against their ‘core’ counterparts.