An Experimental Study on the Learning of Arbitrary and Non-Arbitrary Gender of Pseudo Dutch Nouns by Nonnative and Native Speakers of Dutch

In this article we addres the question of whether, and to what exetent, noun gender attribution in languages such as French, German and Dutch can be formulated in terms of - semantic morphonological rules - competing semantic and morphonological cues - arbitrary idiosyncratic features In addressing this question we carried two learning experiments in which adult subjects, native speakers of Dutch and nonnative second language learners of Dutch, had to learn pseudo Dutch nouns. Our attempt was to provide some counter-evidence to falsify some claims Caroll (1989) made. In Caroll's view, L1 learners of French encode noun gender as an inherent feature of the noun in an obligatory fashion whereas English learners of French learn nouns as words distinct from their determiners. We argue, from the evidence we provide in our study, that the learning of gender values are equally difficult (or easy) for young native speakers, older native speakers, and nonnative speakers when these values are arbitrary. However, when learning the nonarbitrary gender values, it is easier for older native speakers to encode gender of nouns because they have already acquired many nouns from which cues or rules can be derived.