A fundamental question in the study of child language acquisition is that of the relationship between the input the child receives and the mental grammar that the child constructs based on that input. With National Science Foundation support, and under the direction of Dr. Cristina Schmitt, Ms. Karen Miller will investigate the role of variable input on the acquisition of plurality in the Spanish of Chile and of Mexico City. This investigation addresses two main research questions: (i) Can ambiguous input caused by variation affect comprehension while leaving surface production patterns similar to the adult? (ii) How much ambiguity in the input will cause a child to misanalyze the adult input? Adult speakers of Chilean Spanish have a tendency to aspirate ([h]) and/or omit ([zero]) final /s/. For this reason, the plural morpheme is realized sometimes as [zero] and sometimes as [s] or [h], which implies that the Chilean child receives ambiguous input. In other words, the child hears evidence both for plural morphology (the [s] or [h] pronunciations) and against plural morphology (the [zero] pronunciation). On the other hand, in the Spanish of Mexico City, no such reduction or omission occurs and hence the Mexican child receives unambiguous input for a plural morpheme; it is always produced (the [s] pronunciation). This variation therefore allows one to test the hypothesis that children's early grammars are affected by ambiguity in the input they receive. This study uses free speech samples, repetition and naming tasks and various comprehension tasks to test the degree to which children (i) mirror the linguistic variation of the adult population and (ii) understand plural morphology.
The results of this research will have an important social impact on matters concerning education and language testing in children from socially disadvantaged groups, since many tests of early children's language abilities do not adequately take social variation or its effects on learning into consideration.
