Identifying referents for anaphoric nouns and pronouns in spoken English

Abstract

Three experiments were performed in order to test the claim that the difference between the effectiveness of pronouns and repeated nouns in identifying the referent for the listener is affected by (1) the communicative norms shared by the speaker and the listener (2) the distance between the antecedent and the anaphor (3) the status of the antecedent. The results of these experiments seem to provide evidential support for this claim. In Experiment 1, the subjects were presented with sentences in which (1) no intervening sentences separated the antecedent and the anaphor (2) the antecedent was either surface subject or surface object (3) the anaphoric form was either the pronoun or the repeated noun. The results of this experiment showed that when the antecedent is subject, the pronoun was faster than the noun whereas when the antecedent is object, the pronoun and the noun were alike. Experiment 2 differed from experiment 1 in that (1) the antecedent was surface subject only (2) the referent is uniquely identified by the referring expression (e.g. Mrs. Thatcher .... She / Mrs. Thatcher). The results of this second experiment were the same as those of the first (antecedent subject). These results suggest that the difference between the pronoun and the noun (antecedent subject, no intervening sentences) should be viewed as a difference between a unmarked form (ie the pronoun) and a marked one (ie the repeated noun) and not, as it is widely believed, as a difference between an unambiguous form (ie the pronoun) and an ambiguous one (ie the repeated noun). Experiment 3 differed from Experiment 1 in that intervening sentences containing other referents separated the antecedent and the anaphor. The results of this experiment showed that, when teh subject and the object antecedents are treated together, the referent was identified faster when the noun is repeated than when it is pronominalised. Furthermore, Experiment 3 showed that the difference between the pronoun and the noun was affected by whether the antecedent is surface subject or surface object. When the antecedent is subject, the pronoun and the noun were alike whereas when the antecedent is object, the noun was faster than the pronoun. Another aim of Experiment 1 was to compare the non-native speakers' performance with that of the native speakers . This comparison showed that the difference between the pronoun and the noun for the non-native speakers was not the same as that between the pronoun and the noun for the native speakers. The implications of this for teaching the two anaphoric forms to non-native speakers of English were discussed.