The present study investigates the autonomous contribution of syntax and pragmatic to the acquisition of the Hebrew definite article. Hebrew, unlike many Germanic and Romance languages, has only a definite article and does not mark indefiniteness. Moreover, non-discourse related definite nouns, like the sun or the moon, do not necessarily require a definite article. Thus, Hebrew speaking children might get conflicting input on the content of D, and the syntactically mandatory use of the definite article, leading to a later acquisition of the syntactic aspects of definiteness vs. the pragmatic ones. Specifically, we show that children’s knowledge of the pragmatic concept of non-shared knowledge (Scheffer 1997), and the syntactic properties of D, are acquired separately.

Scheffer (1997) argues that the acquisition of the concept of non-shared knowledge yields the acquisition of the properties of D. She argues that additions of definite articles in referential indefinite contexts indicate lack of pragmatic knowledge, whereas dropping of the article in referential (definite and indefinite) contexts is due to lack of syntactic knowledge. Our study shows how the levels of referentiality, marked as syntactic features of D, and the concept of non-shared knowledge interact in licensing the use of the definite article in Hebrew, suggesting that pragmatic principles and syntactic knowledge are acquired separately.
In our study we presented 40 Hebrew speaking children aged 2-5 with an elicited production task which checks their use of the definite article in definite discourse related and non-discourse related contexts and in indefinite referential and non-referential contexts. Children of all age groups used the definite article. There were hardly any additions of the definite article. The ones that were found, however, were in referential context and never in the non-referential contexts and were limited to the twos. These findings support the assumption of early acquisition of the pragmatic principle of non-shared knowledge.

There was also some dropping of the definite article in discourse related definite contexts, as late as the age of four. These results are different from what has been found for English, where the system stabilizes both pragmatically and syntactically by the age of three. The relatively late acquisition of the obligatoriness of the definite article, suggests that pragmatic knowledge which is acquired by the age of three and syntactic knowledge which is mastered in Hebrew only by the age of five (Zur 1983) are two separate modules.

More dropping of the definite article are found with non-discourse related definite nouns like the sun. These findings seem to be related to lack of world knowledge. It is only by the age of five that children develop an intentionality understanding and start to describe beliefs (Astington 1993). Thus, the development of the article system teaches us not only about the separation of pragmatic principles from syntactic ones, but also about the separation of linguistically related pragmatic principles like the concept of non-shared knowledge from cognitively related ones.