Heim
 
Nettverk
 
Korpus/Database
 
Aktivitetar
Prosjekt (pl)
Stormøte 2010
Stormøte 2009
Stormøte 2008
Stormøte 2007
Stormøte 2006
Archive
Program
Workshop
Stormøte 2005
Oslo PhD Course
Arbeidsgrupper
 
NORMS
 
N'CLAV
 
Media
 
Bibliografi
 
Ressursar
 
Blogg
 
[ Logg inn ]
Printer friendly

Solf workshop (ScanDiaSyn 2006 Grand Meeting)

V-to-I as remnant vP-to-SpecTP

Kristine Bentzen, University of Tromsø

It is generally assumed that Norwegian does not have verb movement in non-V2 contexts such as e.g. embedded clauses (modulo embedded V2). This is illustrated in (1), where all the verbs have to follow all adverbs, regardless of the order in which they are merged (the numbers in parentheses indicate the order of merge):

(1) ... ettersom hun alltid(2) måtte(1) kjøpe(3) de dyreste skoene.
as she always must buy the expensive.est shoes.the

Indeed there are dialects in which this is the only possible word order in embedded clauses, e.g. Eastern Norwegian (EastN, spoken in the Oslo area). However, in Regional Northern Norwegian (ReNN) verb movement across outscoping adverbs is possible:

(2) ... ettersom hun kjøpte(2) alltid(1) de dyreste skoene.
as she bought always the expensive.est shoes.the

This is accounted for within a remnant movement approach, and I discuss two potential analyses of the data. The first alternative is inspired by Nilsen (2003) and Müller (2004), who propose that verb second in main clauses involves movement of a remnant projection containing only one specifier and the finite verb to the clause-initial position. In a similar fashion one might suggest that the verb movement exhibited in (2) involves movement of a remnant projection containing only the subject and the verb to e.g. SpecTP.

A second alternative suggests that the subject and the verb have moved independently of each other, rather than together. Then one might assume that the remnant projection moving to SpecTP in (2) only contains the verb (cf. Mahajan 2003), and that the subject has moved on its own to some higher Spec-position. I present some further data from ReNN, concerning subject readings and binding of possessive pronouns, indicating that the latter analysis is more plausible for verb movement in Norwegian embedded (non-V2) contexts.


Det humanistiske fakultet, Universitetet i Tromsø, 9037 Tromsø TLF: 776 44240
Oppdatert av forskar Øystein A. Vangsnes den 26.05.2006 15:46
Ansvarlig redaktør: fakultetsdirektør Jørgen Fossland


Read this page in: Bokmål