Fróðskaparrit - 01.01.1964, Page 104

112
Toward the phonetic description of Faroese vowels
It is no wonder that the short vowels show considerably
more variation in F2 than the long ones. This is due, of
course, to the influence of (especially) following consonants,
which cannot exert the same influence on the initial quality
of long vowels. Moreover, the environments of the short
vowels are in these words more varied than those of the
long vowels. The kind of variation found in the short
vowels is as might be expected in “lax” vowels, but it
should be noticed that some instances of the short vowels
are quite similar to the long vowels. In Fig. 2 the two sets
agree completely with respect to F2, and there is no clear
indication of a basic difference between “tense” and “lax”
articulation (all of this applies to stressed vowels only).
As far as Fl is concerned, both Fig. 2 and Fig. 4 show
that [i] and [e:] (init. part) coincide, and similarly for [s]
and the initial part of [sa]. According to the data shown
in Figs. 2 and 3 there are 3 clearly distinct degrees of
FTraising in long as well as short vowels, but when
these two systems are projected on each other, the result
is a distinction of four degrees, since each short vowel
represents a higher step in the series than the initial phase
of the long vowel with which it is grammatically (ety*
mologically) associated:
(1) i= -
(2) e: 7
(3) sa £
(4) - a
There is no evidence in favour of making a basic distinc*
tion between [e:] and [i] (not even if F3 is taken in ac*
count: F3 of the two vowels is located at 2780 and 2700 cps.,
respectively, as against 2950 cps. for [i:], all average values).
This, of course, does not necessarily imply that [e:] and [i]
sound alike in quality. Taken as a whole, the long vowel
is probably not normally identical with the short vowel,
and the ear may well perceive a difference in quality (as