What for and V2
Øystein Nilsen
University of Utrecht, NORMS/CASTL
What the hell type wh-phrases seem to enforce splitting of so-called what for constructions in Norwegian, while in Swiss German, it does not (Leu 2005):
(1) *Hva faen for en bok har du lest? (No)
what the-devil for a book have you read
(2) Hva (faen) har du lest for en bok? (No)
what (the-devil) have you read for a book
(3) Was zum tuefl bischt du fuer en esel!? (SG)
what to-the devil are you for a donkey
(4) Was zum tuefl fuer en esel bischt du? (SG)
what to-the devil for a donkey are you
In Swiss German, (3) gets an idiomatic reading, roughly meaning: "How can you be so stupid?" while (4) only gets a literal reading. Norwegian seems to allow idiomatic readings both split and non-split versions:
(5) Hva er du for en stut? (No)
what are you for a male.pig
(6) Hva for en stut er du? (No)
what for a male.pig are you
Leu argues convincingly that sentences like (4) are, in fact, instances of what-for split, and hence, instances of V3. The differences with Norwegian seem to indicate that this kind of V3 is forbidden in this language.
This recalls another kind of V3 which is possible in Norwegian, but not in West Germanic (including Swiss German) (Nilsen 2003), only the order corresponding to (8) is possible in these languages:
(7) Jens nesten graat. (No)
John almost cried
(8) Jens graat nesten. (No)
I will argue that these differences show that V2, though superficially similar accross the Germanic languages, does not constitute a unitary phenomenon, and extend the analysis of V2 proposed in Nilsen (2002, 2003) to cover such differences.
