0 Inheritance and motivation in Construction Morphology

patterns can also be concluded from the fact that there is a positive correlation between growth of one’s individual lexicon and acquisition of morphological knowledge. For instance, Smedts (1979) investigated the word formation competence of 13-year-old Flemish Dutchspeaking children, and concluded that this competence is still restricted at that age compared to that of adult language users, and that this correlates with not yet fully developed knowledge of the lexical norm (the set of conventional, established complex words) of one’s native language. The only word formation process that these children were able to use properly and productively was compounding, the most productive and frequently used word formation process of Dutch (Smedts 1979). This shows that language acquisition in this domain is relatively slow, and depends to a great extent on the growth of one’s mental lexicon. The full entry model is a basic ingredient of the model of the hierarchical lexicon in CM, as developed in Booij (2010), which can be characterized as follows: (i) Constructional schemas specify the predictable information of classes of fully specified existing complex lexical items, and specify how similar new complex words can be coined. (ii) Constructional schemas may dominate subschemas that specify additional or more specific properties of subclasses of lexical items. This model does justice to the fact that knowledge and use of abstract morphological schemas is dependent upon the memorized knowledge of sets of complex words that instantiate these patterns. Therefore, morphological schemas should not be seen as formal mechanisms for achieving maximally parsimonious lexical representations. Instead, they have two other functions: schemas motivate the existence of the relevant set of complex words, and they predict how this set can be extended (cf. Goldberg 2006 for the same point for syntactic constructions). The motivating function of schemas has the effect of reducing the degree of arbitrariness of formmeaning relations in complex words in the lexicon. Schemas also structure the lexicon. For instance, the schema for English deverbal nouns in -er in (1) defines a word family with shared properties, and thus provides a partial structure to the English lexicon. The full entry theory is also defended in Jackendoff (1975; 1997). Jackendoff (1997) concluded that given a full entry theory, schemas have to play a role in computing the