Íslenskt mál og almenn málfræði - 01.01.2000, Blaðsíða 92

90 Eiríkur Rögnvaldsson
found in 16th century Icelandic, but they have disappeared completely from all the
Modem Scandinavian languages. Since these different imperative constractions
disappear at the same time, it seems reasonable to assume that there is only one
change involved, not two — one in main clauses and another in subordinate clauses.
The change in question is that imperative verbs must stand in C in Modem
Icelandic, whereas they did not necessarily have to do so in Old Icelandic. I suggest
that the reason for this is that the imperative operator was optional in imperative claus-
es in Old Scandinavian, whereas it has now become obligatory. When the imperative
operator was not present, nothing prevented subjects and other phrases from moving
to Spec-CP in main clauses; and the imperative verb could stay in 1° in subordinate
clauses. It is of course usually assumed that the relevant operators must be present in
yeí/no-questions, imperatives, Narrative Inversion, etc., in order to ensure the correct
interpretation of such sentences. However, imperative clauses differ from questions
and sentences with Narrative Inversion in one important feature; they carry a morp-
hologically distinctive marker, the imperative form, that the other types do not have.
In languages having a morphologically distinctive imperative form, therefore, we do
not need an imperative operator in order to interpret imperative clauses as commands,
because the form of the verb will ensure such an interpretation anyway.
In imperative main clauses in Old Icelandic, the verb must stand in the initial
position in non-conjoined clauses and after og, whereas it can (and usually does) fol-
low the subject in clauses beginning with en. If the word order in imperative clauses
were an independent syntactic phenomenon, conditioned by the presence of an impera-
tive operator in Spec-CP but unrelated to other word order factors, we would expect
the examples of subject-initial imperative sentences to be evenly distributed among
these three types of main clauses. The fact that they are not seems to indicate that the
word order in imperative main clauses did, to some extent at least, obey the same
principles as the word order in declarative clauses; that is, it was in part pragmatically
conditioned. Thus, the change in the syntactic position of imperative verbs ffom Old
to Modem Icelandic can be interpreted as a case of grammaticalization, since the posi-
tion of imperative verbs is no longer dependent on semantic or pragmatic factors in
any way, but instead govemed by purely syntactic factors.
Eiríkur Rögnvaldsson
Háskóla Islands
Arnagarði við Suöurgötu
IS-101 Reykjavík, ÍSLAND
eirikur@hi.is