摘要：This paper aim to explain why the prenominal definiteness marker found in modified structures only is acquired much later than the suffixal definite article in Norwegian. The coexistence of the two definiteness markers is the result of the so-called double definiteness phenomenon in Norwegian which occurs in definite structures involving an attributive adjective. As the prenominal determiner only occurs in modified contexts, one obvious explanation that presents itself is that the omission of the prenominal determiner is related to the fact that it is much less frequent in the input than the suffixal article. However, if we consider the form of this determiner, it is clear that the form itself must be very frequent in the input, as it is homophonous with the demonstrative and with inanimate third person pronouns, which are all frequent in use. Consequently, a lexical insertion approach to the double definiteness phenomenon is proposed according to which the discrepancy in the order of acquisition is argued to be due to a combination of prosodic saliency and the lexicalisation of semantic features, the latter of which will be dealt with here.
