Íslenskt mál og almenn málfræði - 01.01.1992, Page 34

32
Ari Páll Kristinsson
Canadian Journal ofLinguistics—Revue canadienne de Linguistique 30,4:415—
451.
Sigríður Sigurjónsdóttir. 1988-1989. Nokkur orð um aðblástur, fráblástur og afröddun
í íslensku. íslenskt mál 10-11:130-137.
Snorra Edda. 1975. Ámi Bjömsson bjó til prentunar. Iðunn, Reykjavík.
Vennemann, Theo. 1972. On the Theory of Syllabic Phonology. Linguistische Berichte
18:1-18.
Wessén, Elias. 1945. Svensk sprákhistoria I. Ljudlara och ordböjningslara. 2. útgáfa.
Stokkhólmi.
SUMMARY
The aim of this article is to account for the epenthetic u in Icelandic, i.e. the historical
change when the vowel u occurred after a consonant followed by either a word-final r or
an r which was in tum immediately followed by another (tme) consonant (fiskr became
fiskur.fegrð became fegurð, etc.). It is maintained here (cf. Ari Páll Kristinsson 1987)
that w-epenthesis took place from the last quarter of the 13th century to the first quarter
of the 16th century, at least, and that it was optional in that period. (Traditionally,
the change is considered to have been completely over toward the end of the 14th
century.) Certain types of phonological environment, namely when r stood between
other consonants (i.e. tme consonants, not between a consonant and a glide) or when r
was word-final but preceded by one of the hard stops, p, t or k, called, however, more
often for w-epenthesis than others.
It is found unlikely here, referring to (universal?) sonority hierarchies and strength
hierarchies, that syllables e.g. of the fiskr or fegrð type could have preserved a non-
syllabic r from the time of the Proto-Nordic syncope to the end of the 13th century,
when the earliest examples of an epenthetic u appear in Old-Icelandic manuscripts.
Therefore, it is proposed that a syllabic r optionally developed in the phonological
environment under consideration here, offering speakers some relief in pronunciation,
especially the pronunciation of clusters that would otherwise be „difficult", such as
those in words of the grípr, bátr, tekr or fegrð types. But toward the end of the 13th
celitury the change starts developing that this syllabic r could be replaced as an option in
pronunciation by an epenthetic u followed by an r which in tum became automatically
non-syllabic.
The Icelandic M-epenthesis is described here along these lines using non-linear and
autosegmental formalism. (5), (6) and (7) describe syllables at different stages in the
progress. (8) is a mle necessary to tum a post-consonantal r, which was either word-
final or pre-consonantal, into a syllabic and thus create a new syllable, and (9) shows
the kind of „breaking" of place features of the syllabic r and the reorganising of the
CV-tier, known as w-epenthesis. (9) shows that all laryngeal and manner features are
common to w and r. (Áhersluliður = foot, atkvæði = syllable, stuðull = onset, rím =
rhyme, kjami = nucleus, kálfur = coda, radd. = voiced, sp.rg. = spread glottis, samf. =