Íslenskt mál og almenn málfræði - 01.01.1997, Page 182

180
Þórhallur Eyþórsson
SUMMARY
The objective of this paper is to investigate how old the verb second phenomenon (V2)
is in Germanic, examining texts in Runic Norse (Northwest Germanic), Old Norse-
Icelandic, Old High German, Old Saxon, Old English and Gothic. I show that contrary
to the standard account (Weerman 1989; Kiparsky 1995) already at the earliest attested
stage — in the language of the runic inscriptions and Gothic — there is evidence sup-
porting the assumption of movement of the finite verb from its base in VP to positions
higher up in the clause. I propose that all old Germanic languages, including Gothic,
exhibit V-to-C movement in direct questions, with fronted negatives and in commands,
as well as in some types of verb-initial declarative main clauses.
On the other hand, the widely held view (due to van Kemenade 1987) that in Old
English topicalized complements involve obligatory verb movement is contradicted
by various facts. Topicalized complements do not seem to involve verb movement in
Gothic either. In this respect, these languages are like Modern English, whereas other
old and modern Germanic languages exhibit obligatory V-to-C movement in topical-
izations. Moreover, an investigation of runic inscriptions indicates that verb-final
clauses in this text corpus are fewer than is usually claimed. Surprisingly, topicalized
complements may have involved movement of the verb to C already at the earliest
stage in the northern part of the Germanic linguistic area. The distribution of V2 in
Old Germanic suggests that is not the outcome of a single rule, but rather of various
independent rules conspiring to target particular structural positions. Finally, I
speculate that the fmdings of this paper may have interesting consequences for the
evaluation of the dialectal situation in Old Germanic.
Þórhallur Eyþórsson
Department of Linguistics
University of Manchester
Manchester M13 9PL
United Kingdom
tolli@man.ac.uk