| dc.description.abstract | The issue of inflectional morphology feeding word formation has occupied the minds of many linguists for decades. A particular example of this phenomenon is regular plural inflection inside Dutch complex words. With respect to this phenomenon, the theories of inherent versus contextual inflection, linking morphemes, stem allomorphy and sets-of-subsets are discussed, where the first and last theory are semantic in nature, and the second and third are largely morpho-phonological. However, these theories cannot satisfactorily explain some of the issues regarding why and how regular plural inflection may appear inside Dutch complex words.
An alternative account, the collective-versus-generic-interpretation theory, is proposed to account for regular plural inflection inside Dutch complex words. The central idea in this semantic theory is that when non-head nouns seem to be regularly inflected with an -en plural suffix, these are in fact noun plurals with either a generic or a collective reading. In the case of a generic reading (determined by the head), the [+plural] NUMBER information of the non-head is suppressed, but still very much a plural (hence the orthographic appearance of a plural). In the case of a collective reading, the -en noun plural does trigger [+plural] NUMBER semantics, and it can be unequivocally established that inflection indeed may feed derivation. However, this theory only accounts for -en plurals inside Dutch complex words, as -s plurals almost never appear inside Dutch complex words (the exception being certain pluralised diminutives). When an -s morpheme appears inside compounds and derived words, it is there generally to function as a linker for phonological purposes, or to denote a genitive.
In order to test the predictions made by the theory, an offline experiment was conducted where bare nouns and non-head nouns inside complex words were suffixed with an incorrect regular plural morpheme. The assumption here is that if plural NUMBER semantics are already suppressed inside complex words with a generic non-head, then the incorrect suffixation of a wrong plural morpheme is less salient. On the other hand, if bare nouns are incorrectly suffixed with a wrong plural morpheme this will be more salient as the [+plural] NUMBER status of the noun plural is switched to a [-plural] NUMBER status. The results of this experiment show that although both suffixations are considered ungrammatical by native speakers of Dutch, the incorrect suffixations in the former scenario are considered less ungrammatical than in the latter. Hence, the predictions of the collective-versus-generic-interpretation theory are borne out. | |
