William Auguste Foley

This project investigates the question whether there are innate cognitive biases that universally determine word classes in language. This will be investigated in detail in the Austronesian language family, where the status of the noun-verb distinction has long been controversial. We will approach the problem ontogenetically, by a systematic study of a representative sample to establish what evidence there is for coherent classes of nouns and verbs, but, more importantly, in light of a co-evolutionary account, phylogenetically, by how the patterns of words have shifted over time. Austronesian languages exhibit a high degree of diversity: sometimes a tendency to distinguish nouns and verbs seems clear, but elsewhere it seems extremely problematic. In most language families, the predisposition to distinguish nouns and verbs is strong. So what is it about Austronesian languages that leads to a recurring predilection to attenuate the distinction? Given the widespread attestation of a robust noun-verb distinction elsewhere, its attenuation is an analogue of a recessive trait. What conditions select for or against this? What are recurring solutions, evidently co-evolutionarily favoured, and what are outliers, statistically low probability outcomes, and why? While at first glance seemingly quite specialized, the project has broad implications for cognitive and anthropological science. It is the first detailed study within a co-evolutionary framework of any major grammatical property. It seeks to understand linguistic structure and its possible changes through the selective interplay between cognitive biases due to likely panhuman biological givens and specific historical developments. This project probes some of the limitations on innate cognitive endowment for the formation of linguistic structures and their transitions through time and space, in particular the role that language-specific, in a word, cultural, learning plays in language acquisition.