An elicited production test of the optional infinitive stage in child Spanish

It has been suggested that Spanish-type languages do not pass through the Optional Infinitive (OI) stage (e.g. Guasti 1994, Wexler 1998). Recent work (Pratt and Grinstead 2007b) has challenged this assumption, arguing that this conclusion stems from a failure to accurately count bare stems as non-finite verbs. In the current study, both a grammaticality choice task and an elicited production task of verbal finiteness are carried out with typically-developing children as well as children with SLI in order to crossvalidate the earlier child Spanish OI finding with a second methodology. In the elicited production task, it is shown that typically developing children (n=30, ages 3;5-6;5, mean age = 5;1) correctly mark verbs for finiteness 86.8% of the time (469/540). The children's scores correlate with their ages (Pearson correlation = .534, p = .002, two-tailed). With another group of typically-developing children (n=22, ages 3;7-6;5, mean age = 5;2), it is shown that their scores on the grammaticality choice task and the elicited production task are correlated (Pearson correlation = .697, p < .000). Finally, it is shown that the scores of 21 Spanish-speaking children with SLI (ages 4;9-6;4, mean age = 5;7) on the elicited production task are significantly worse (for a paired samples t-test, t(20) = -5.794, p < .0001) than those of an age-matched (n=21, mean age = 5;6) typically-developing control group. In summary, these results show that child Spanish-speakers' finiteness judgments on the grammaticality choice task correlate with their elicited productions, validating the findings in Pratt and Grinstead (2007b), and supporting the existence of an Optional Infinitive stage in child Spanish.