| Summary: | Tromsø Norwegian is a variety of Norwegian that is part of the mainland Scandinavian isogloss in which a contrast between two tonal accents, known as accent 1 and accent 2, has developed. The distribution of the two tonal accents in morphologically complex words is subject to both phonological and morphosyntactic factors, which interact in non-trivial ways. This dissertation uses this interaction as a prism to investigate the interface between morphosyntax and phonology. In order to shed light on the issue, the dissertation addresses three important aspects. (i) The relationship between the tonal accents: this dissertation sides with research that only allows lexical marking of accent 1, whereas accent 2 is always post-lexical and thus belongs to phonology proper. (ii) The morphosyntactic structure of what is conventionally labelled as compounds: the dissertation takes the constructivist view on word formation, advocated for in Distributed Morphology, that morphologically complex words such as compounds are created by Syntax and not by the Lexicon. It is also assumed that Syntax is able to assemble objects in parallel in different working spaces, with the consequence that certain morphologically complex structures arise as a result of adjunction. (iii) The nature of the communication between morphosyntax and phonology: it is assumed that the communication between the two modules of language is carried out derivationally while respecting modularity, as advocated by Phase Theory. The analysis that is offered shows that domains for assignment of tonal accent in compound structures fall out from the morphosyntactic derivation as phases, which indicates that there is no need to assume any representational device in phonology for these objects. The analysis also has important consequences for the nature of the tonal accents. More specifically, it is revealed that the non-lexical manifestations of the tonal accents are best seen as expressions of uncategorised metrical structure rather than being tone proper: accent 1 ...
|
|---|