De Relatie Tussen Theoretische en Toegepaste Taalkunde

Research in the field of language acquisition has had surprisingly little influence on linguistic theory. In part this might be due to the fact that applied research is empirically more complex given the instability of various stages and the difficulty in interpreting the data. Moreover, applied linguists are generally more interested in explaining the data than in contributing to the development of the theory. In contrast, theoretical linguists often take data and analyses from the domain of language acquisition seriously to the extent that these confirm their theory. In this paper it is argued that a long-standing problem in linguistic theory can be solved if we take the generalizations from acquisition into account. The empirical issue concerns the problem of separable compound verbs in Dutch. The theoretical issue involved concerns the definition of the notion 'word'. From a survey of the rather extensive literature on this problem, it is clear that separable compound verbs may be analysed either as single words or as complexes consisting of two separate words. It is demonstrated that language acquisition data provide us with strong arguments in favour of an analysis in which separable compound verbs are syntactically complex.