The interplay between classifier choice and animacy in Mandarin-Chinese noun phrases: an ERP study

Animacy is an important semantic assignment principle in both gender and numeral classifier systems. Linguistic research has shown that animacy is not a simple binary feature but represents a fine-grained taxonomy of different animacy levels. We used the classifier system of Mandarin Chinese, with classifiers varying in semantic constraint, to assess whether the referents' degree of animacy influences the processing of classifier-noun pairs. ERP results show an effect of agreement mismatch only when the general classifier (sic) ge is paired with nouns referring to higher animals (chimpanzee, lion). For these pairs, the sortal classifier (sic) zhi has to be chosen. No such effect was observed for nouns that refer to intermediate-level and lower animals (octopus, earthworm), indicating that (sic) ge does not constitute an agreement mismatch here. We also observed significant ERP effects indicating that speakers of Mandarin Chinese process the general classifier and the specific sortal classifiers differently.