On Mon, 31 Mar 2008 20:57:36 -0500, "Patrick Ryan"
<
proto-language@...> wrote:
>To undergo Umlaut or not to undergo Umlaut is a question of whether a
>language anticipates following vowel qualities by quality changes in the
>immediate vowel.
>
>That Dutch allows Umlaut with short vowels but no long vowels is interesting
>but has no bearing on our question.
>
>Emphatic coloration, as does actually exist in languages like Arabic, is a
>vowel quality change brought about by an adjacently _preceding_ (not
>_following_ as above) oral configuration.
Emphatic coloration (like laryngeal coloration in PIE) works
both ways (Eg. Arabic s.Aff "row", bAt.n "stomach", fÁd.d.A
"he emptied").
>A corollary is that the agent and
>the patient are always in the same syllable which is never the cause with
>Umlaut.
>
>So, it is perfectly natural that any quality change effected by an emphatic
>would affect but short vowels and long vowels in the same syllable.
Of course it's perfectly natural that colouring would affect
short and long vowels alike. It's also perfectly natural
that it would affect short vowels more than long vowels.
Long vowels are simply "stronger" than short vowels, just
like geminate consonants are "stronger" than single ones.
Another example: Old English breaking (fracture) affects
front vowels before "velars" (h, r, l, w?). Most affected is
/æ/, then /e/, then /i/. Even less affected (but still
affected) are the long vowels /æ:/ and /i:/. The hierarchy
is short/low - short/high - long/low - long/high.
See
https://dspace.ucalgary.ca/bitstream/1880/25119/1/38538Kitch.pdf
p. 111-112 (Conclusions):
To characterize the parameter which indicates vowels in
breaking, I incorporated the properties of height and
length, which are reflected to varying degrees in breaking
diphthongs. In consideration of the differing strength
values of low, mid and high vowels, according to Vennemann
(1 988), and also, that long vowels typically demonstrate
more resistance to vowel mutation than short vowels, as
outlined by Goldsmith (1990), I suggested that the vowels
involved in breaking could be graded within a scale
according to their intrinsic strength, or the degree of
resistance to diphthong formation they demonstrate
in breaking. This vocalic parameter was labelled as vowel
integrity.
=======================
Miguel Carrasquer Vidal
miguelc@...