Íslenskt mál og almenn málfræði - 01.01.1984, Qupperneq 138

136 Kristján Arnason
be placed at all. A salient feature of Modern Icelandic phonological
patterning is that monosyllables may be formed from bisyllabics by
deletion of the second syllable, and conversely a disyllabic can be
formed by adding an appropriate ending to a monosyllable without
any sort of restructuring being visible in the remaining (or original)
monosyllable. Thus, the same sort of structure prevails in the mono-
syllable hest [hes$] as in the first syllable of hestur [hes^Yr]. This seems
to be a general feature of the phonology; pairs like vors : vorsins
[vojs], ‘spring’ (gen. sg. indefinite vs. definite); gelt : gelta [íe]^] ‘a
bark’ vs. ‘to bark’; diskur : diskó : disk [$lsg] ‘a plate’ (nom.) vs.
‘discoteque’ vs. ‘a plate’ (acc.); œskti, æskt [ais(g)<£| or [ais(x)$], ‘to
wish’ (past vs. past participle); bölva : bölv [þœlv] ‘to swear’ vs. ‘the
act of swearing’, show, as far as I can judge, no difference in the
structure of the stressed syllables. This suggests that a case can be
made for looking at the invariant forms in the respective pairs listed
above as units in the phonological structure of Icelandic. This would
entail syllabication of the sort hest.ur, æskt.i, bölv.a etc.
We may note that this sort of syllabication is assumed to be phono-
logically motivated but it does not derive from direct phonetic support
by way of e.g. experiments showing regularities in speech production
or speech perception that would unambiguously entail the above syl-
labication.
It seems also that it is safe to maintain that the motivation is not
morphologically based (and then no motivation for the phonological
syllable), although the pattern observed is heavily made use of in the
morphosyntax (most of the second syllables being representatives of
morphological units as inflectional endings). The phonological char-
acter of the motivation is reflected in the fact that the structures oc-
curring in positions after the first syllable regularly show different
phonological patterns from those characteristic of the initial syllables.
Firstly, it may be mentioned that lexical primary stress falls on initial
syllables, and the phonological systems available to the unstressed syl-
lables are much more restricted than those available to the stressed
ones: there are fewer vowel qualities (the most common are [a,l,Y])
and length is excluded. Another phonological phenomenon that seems
to have a bearing on this is the fact that the length of the stressed
vocalism, which is regularly assumed to have two degrees, long and