Danish grammatical tradition has it that V2 and V3 word order are signals of the syntactic
difference between main and subordinate clauses. However, several studies have shown that
both word orders appear in both clause types (Heltoft 2005, Christensen 2007, Jensen forthc.).
Though V2 and V3 are asymmetrically distributed, we argue that the word order difference
should rather be seen as a signal of (subtle) semantic differences. In main clauses, V3 is
highly marked in comparison to V2, and occurs in what may be called emotives. In
subordinate clauses, V2 is marked and signals what has been called ”assertiveness”, but is
rather a question of foregrounding (cf. Simons 2007: Main Point of Utterance).
The paper presents the results of a study of word order in subordinate clauses in contemporary
spoken Danish and focuses on how to include the proposed semantic difference as a factor
influencing the choice of one variant over another in a (socio)linguistic variable. This is a
crucial methodological issue in the study of syntactic variation since variants are hardly ever
semantically equivalent in all respects. The study, which is a part of the LANCHART project
(www.lanchart.hum.ku.dk), is based on panel studies of two age cohorts of speakers in
Copenhagen, recorded in the 1980s and again in 2005-07, and on recent recordings with two
age cohorts of speakers from the western part of Jutland. This makes it possible to study
variation and change with respect to word order in subordinate clauses in both real and
apparent time, as well as geographical variation.
The results show that V2 word order in subordinate clauses is much more frequent than
commonly assumed. Furthermore, they indicate that the most decisive factors predicting word
order in subordinate clauses are the syntactic function of the clause and the type of
subordinating conjunction, although social and geographical factors also have an impact. The
results are consistent with the hypothesis that V2 word order is associated with foreground or
main point of utterance, if we accept it as a statistical tendency in language use rather than as
an invariant coding.

References
Christensen, T. K. 2007. Hyperparadigmer – en undersøgelse af paradigmatiske samspil i
danske modussystemer. Institut for kultur og identitet. Roskilde, Roskilde
Universitetscenter. PhD thesis.
Heltoft, L. 2005. Ledsætning og letled i dansk. OV-sætningens rester. L. Heltoft, J. Nørgaard-
Sørensen & L. Schøsler. Grammatikalisering og struktur. København, Museum
Tusculanum.
Jensen, Torben Juel forthc. Ordstilling i ledsætninger i moderne dansk talesprog. Ny forskning
i grammatik.
Simons, M. 2007. Observations on embedding verbs, evidentiality and presupposition. Lingua
117 (6), 1034-1056.
Vikner, S. 1995. Verb movement and expletive subjects in the Germanic languages. Oxford
University Press.