|
Workshop on Inversion and Verb Movement
Speculations on the interaction of V-to-I and embedded V2
Gunnar Hrafn Hrafnbjargarson
University of Tromsø, NORMS/CASTL
So far, we have assumed that V-to-I movement and V2 are two independent features that do not interfere with each other. The argument in favor of such an assumption comes from Mainland Scandinavian and Faroese which all have main clause V2, but none have V-to-I. Furthermore, none of the Norwegian and Faroese dialects that have some kind of verb movement in embedded clauses have V2 (except in the context of a bridge verb).
Icelandic has V-to-I movement, in both main and embedded clauses, and V2 is possible in almost all types of embedded clauses. Few exceptions to the availability of embedded V2 are control infinitives, questions and relative clauses, which might be interpreted as an argument in favor of what I will call the "no connection"-hypothesis, i.e. that embedded V2 and V-to-I movement may not have anything to do with each other. However, phenomena that have previously been related to V2, such as stylistic fronting, only occur in embedded clauses where V-to-I movement is possible and where V2 is impossible (control infinitives, questions and relative clauses).
In many ways, stylistic fronting resembles topicalization and I have in a previous analysis proposed that stylistic fronting should be analyzed as an instance of movement into the CP-layer. Stylistic fronting has also been related to the presence of V-to-I movement and stylistic fronting arguably plays a role in the loss of V-to-I movement in Mainland Scandinavian. If assumed that stylistic fronting depends on V-to-I movement and that it can be analyzed in a similar way as V2, the question arises whether embedded V2 could also be related to the presence of V-to-I movement.
In my talk, I will take the discussion one step further and speculate on the idea whether there might be a connection between the presence of V-to-I movement and availability of embedded V2.
|
