Please join us this Thursday at 4:15 for this practice BU talk:

Intervention effects in Mandarin-speaking children’s comprehension of passives 

Minqi Liu, Victoria Mateu, and Nina Hyams

In Mandarin, passives with an external argument (EA) – “long passives” (1a) – are structurally different from “short passives” (1b), in which the EA is not projected (e.g., Huang 1999). Because the EA structurally intervenes in the dependency between the surface subject and its gap in long passives (2a) but not short passives (2b), the Intervention Hypothesis (e.g., Friedmann et al. 2009) predicts that (i) Mandarin long passives should be harder for children to comprehend – an Intervention Effect – and (ii) this Intervention Effect should decrease when the morphosyntactic features of the two arguments in long passives are distinct.

We tested 80 monolingual Mandarin-speaking 3- to 6-year-olds (M = 4;11) in a sentence-picture matching experiment with three sentence types (Actives, Long passives, and Short Passives) and found that long passives were the most difficult for children’s comprehension.

Furthermore, in this experiment we manipulated the classifiers of the argument DPs in order to examine the effect of featural match vs. mismatch between the two arguments in actives and long passives, and to ask what kind of features is relevant in intervention. In particular, we tested sentences containing one or two of these three classifiers: (i) a [PLURAL] classifier –xie that encodes the grammatical Number feature, (ii) a shape-specific classifier -tiao that s(emantically)-selects for long-bodied entities such as snakes and fish, and (iii) a general classifier –ge that is non-plural and not shape-specific. Data showed that a mismatch in Number (-xie vs. –ge), but not lexical properties such as body shape (-tiao vs. –ge), between the two arguments improves children’s comprehension of long passives, but not actives.

These results are consistent with the predictions of the Intervention Hypothesis and support grammatical accounts to Intervention such as (featural) Relativized Minimality (e.g., Rizzi 1990, 2004).