Íslenskt mál og almenn málfræði - 01.01.1985, Blaðsíða 142

140
Christer Platzack
classified like this in all the three texts, whereas mæla and svara are
NON-VS-Verbs in two texts; furthermore, none of the examples of
these verbs in my /s-sample (3 occurrences of each verb) are found in
the VS-construction. This is an interesting result, considering the fact
that the only cases of the VS-construction in declarative main clauses
in Modem English are found with verbs of saying, as in the following
example:
(11) Said the Prime Minister: „From now on, we have to rely on ex-
ternal sources“
This use of inversion in English is a historical relict. However, we
should not be surprised to find a use of the VS-pattern in English
which differs from the use of this construction in Old Icelandic: within
certain limits, different languages may choose (for historical reasons) to
use corresponding constructions in different ways.
3. Discussion
In this concluding section, we will consider the VS-construction
from the point of view of grammatical theory. As we have seen,
Kossuth (1978) has taken the relatively high frequency of VS-sentences
in Old Icelandic as an incitement to classify this language as an
SV-VS-language, whereas Sigurðsson (1983) rejects this classification
and argues that Old Icelandic was an SVO-language. However, the
whole debate might well be a pseudo-debate, as Rögnvaldsson
(1982:22 ff.) indicates, and it certainly is one if word order typology
has something to do with the underlying grammatical system. It is to
be noticed that frequency says something of how a certain construction
is used, but it does not tell us anything of the place in the grammatical
system of a certain construction.
Widening our perspective, it seems evident that the VS-construction
is closely related to the verb second constraint found in all modem
Germanic languages except modem English. According to this con-
straint, a declarative main clause, as well as a direct w/z-question, must
have the finite verb in second position. It was claimed already by
Diderichsen (1941) that this constraint (of course, Diderichsen did not
use this term) is due to an underlying VS-order of the nexus-field of the
main clause: according to Diderichsen, the nexus field consists of the