Revisiting the population vs phoneme-inventory correlation

Speculation about the relationship between linguistic and non-linguistic structures dates back at least a century. Sapir (1912) suggested that the infuence of non-linguistic factors (such as topography, climate, fora and fauna, etc) are most clearly refected in a language’s vocabulary, but Sapir also believed that they afect the phonological and grammatical systems of languages. It is clear that certain non-linguistic contexts clearly favor diferential enrichment of the lexicon, evidenced by the uneven distribution of domain-specifc vocabulary in relation to the importance of those domains for diferent linguistic communities (e.g., Nettle, 1999). However, the relationship between phonologies and extralinguistic factors like social structure or population size has been more controversial (e.g., Atkinson, 2011; Bakker, 2004; Hay & Bauer, 2007; Lupyan & Dale, 2010; Nettle, 1999; Pericliev, 2004; Trudgill, 1996, 1997, 2002, 2004; Wichmann & Holman, 2009; Wichmann, Staufer, Schulze, & Holman, 2008).