Íslenskt mál og almenn málfræði - 01.01.2005, Page 103

Ég er, ég vill og ég fær 101
Icelandic to Modem Icelandic shows a tendency for verbs to transfer from Pattem C to
Pattem B.
Pattems A and B are thus characteristic of the singular active paradigm and, as
argued in section 5, they constitute the system-defining stmctural properties of the
singular active in Icelandic, as defined by Wurzel (1989). Wurzel s theory predicts
that pattems that do not agree with the system-defining structural properties are prone
to changes and tend to merge with the more system-congruous pattems. This is pre-
cisely what can be observed in Icelandic. The preterite indicative of weak verbs, the
present conjunctive and preterite conjunctive of all verbs, has moved from Pattem C
to Pattem B in a series of changes that took place mainly in a period from the 14th
century through the 16th century. In Old Icelandic, the present indicative ofvera to
be’ also followed Pattem C, 1 e-m, 2 es-t/er-t, 3 es-0/er-0, but already in the first
half of the 14th century, as a result of the change em —* er in the 1 st person, the para-
digm was transferred to Pattem B, 1 er-0, 2 er-t, 3 er-0. Thus the present paradigm
of vera became system-congmous.
The changes currently underway in the verb vilja are part of this same morpholog-
ical tendency, aiming at maximal system congmity. The standard singular paradigm
of vilja follows Pattem C with its three-way distinction, 1 vil-0, 2 vil-t, 3 vill-0', con-
sequently, it is not system-congruous. System congruity is obtained by each of the
two changes—the change in the lst person vil-0 -* vill-0 or the change in the 3rd
person vill-0 —* vil-0—both of which yield paradigms following Pattem B.
In the first change, the lst person takes up the form of the 3rd person; this is not
surprising as the 3rd person is generally regarded as unmarked. The second change,
however, raises questions, since it involves the lst person form replacing the 3rd per-
son form. This may to some extent be due to hypercorrection, as the lst person form
vill-0 is generally considered substandard; the stigma associated with lst person vill-0
may have triggered the replacement of 3rd person vill-0 by vil-0. Also, as argued by
Tiersma (1982), verbs of perception and emotion tend to be used very frequently in
the lst person; consequently, the lst person of such verbs may be considered locally
unmarked. This, too, may have contributed to the change in the 3rd person of vilja.
The analysis proposed here predicts that other verbs currently under the incongm-
°us Pattem C will eventually be transferred to one of the system-congruous pattems,
A or B. There are, in fact, indications that this prediction may be correct. The written
language on the Intemet shows numerous instances of the lst person form fœr of the
verb/á ‘get’, suggesting an incipient change in the 1 st person,fœ-0 —*fœ-r, by which
Ihe present paradigm 1 fœ-0, 2fœ-rð, 3 fœ-r with Pattem C is replaced by a system-
congmous paradigm with Pattem B: 1 fœ-r, 2 fce-rð, 3fce-r.
/faraldur Bernharðsson
Stofnun Árna Magnússonar
Arnagarði við Suðurgötu
/S-101 Reykjavík, ÍSLAND
haraldr@hi.is