Language diversity from individual-level biases
In this talk I’ll critically revisit the idea that individual-level biases shape language diversity. The idea is straightforward: the distribution of language structures in the world approximately follows the marginal payoff they provide to humans, in the form of ease of production, learnability, expressivity, and the like. I will discuss a number of cases from across linguistic domains but will ultimately focus on the role of learnability (as evidenced in individual-level experiments). In this regard there are multiple related hypotheses, including the notions that language structures which are (1) easier to acquire by children, (2) easier to acquire by adults and (3) easier to retain by adults, are overrepresented across the world’s languages. The main evidence I’ll use is the history and the nature of languages which instantiate almost ideal case studies for each of the hypotheses described above, accordingly: (1) Creole languages (Blasi, Michaelis & Haspelmath 2017 Nat Hum Beh), (2) languages spoken by large populations (Scherbakova et al. to appear in Sci Adv), and (3) surviving languages (i.e. languages that are no longer transmitted to newer generations, Blasi et al. in prep). I will conclude that the evidence for a direct pipeline between individual-level biases and social-level language structures is surprisingly less robust than widely assumed across the cognitive and language sciences, and I will discuss some future research directions.
*Anyone not affiliated with CEU wishing to attend in-person in Budapest must RSVP to get access to the lecture hall.