The Role of Script and Phonology in Lexical Representation

Abstract

Two experiments were conducted to evaluate the role of script in lexical representation in bilinguals. The particular issue under investigation concerned the role of a script difference when the translations concerned are in all other respects identical. Does it provide a basis for the operation of unique functional entries in the lexicon, or is it, like case, treated as a pre-lexical attribute? The experiments contrasted the physically unrelated scripts used to depict Hindi and Urdu, North Indian languages which share a substantial vocabulary and which, with minor qualifications, are phonemically identical. The experiments used repetition effects in lexical decision and memory for language of presentation procedures, respectively, to assess the representational status of items common to the Hindi and Urdu vocabularies. Experiment 1 demonstrated that transfer is virtually complete between Hindi and Urdu translations, suggesting that the role of script is akin to that of case, and quite unlike phonology in its impact on lexical definition. Experiment 2 showed that whereas memory for language of presentation is reliable in tests involving language pairs such as Hindi and English, performance on this task is poor when script alone provides the distinguishing feature. Considered together, the results of the experiments are consistent with the proposition that when translations differ only with regard to script, a common lexical unit is involved in visual recognition and attribute retention. By implication, therefore, a morphophonemic distinction is a pre-requisite for the presence and use of distinct units for translations. The representational character of this interpretation was supported by the fact that complementary evidence was obtained from tasks involving qualitatively different retrieval processes.