Abstract

The use and acquisition of the get-passive has so far yielded a variety of accounts and suggestions. This paper presents new experimental evidence concerning the use and the acquisition of the get-passive by children, as well as adult judgments of get- and be-passives. Within a prototype approach to the passive, experiments investigated when 2–4-year-old British children produce get- as opposed to be-passives. The role of direct affectedness of the patient on
get-passive production was investigated further in a follow-up experiment. In addition to the child data, ratings of get- and be-passives were obtained from British English adult speakers to investigate the acceptability of these passives and their relationship to developmental data. The
first experiment showed that the chosen prototype approach clearly predicts children’s acquisition of be-passives with get-passives being more peripheral members of the category
‘passive’ than be-passives. The second study shows that even if the child herself is the affected patient in the play action, get-passives are only rarely produced. In contrast to American children, direct affectedness did not induce British children to produce a significant amount of getpassives. Last, adult ratings confirm that British English speakers rate be-passives consistently as better examples of passive sentences than get-passives. The evidence suggests that getpassives are more peripheral for British than for American children and adults. Implications for the possible role of parental input and the validity of existing accounts of the get-passive are discussed.