Re: [tied] Re: Indo-Iranian Vowel Collapse (was: IIr 2nd Palatalisa

At 8:50:45 PM on Thursday, November 24, 2005, Rob wrote:

> --- In cybalist@yahoogroups.com, "Brian M. Scott"
> <BMScott@...> wrote:

[...]

>> As I understand it, PWGmc */a/ > early OE /æ/; before /lC/,
>> /rC/, and /x/ this broke to /æa/ (spelled <ea>) in West
>> Saxon and retracted to /a/ (which was probably more like
>> [A]) in Anglian. (When the conditions for breaking weren't
>> present, /æ/ remained /æ/ except in Kentish in southwest
>> Mercian, where it became /e/.) Finally, short vowels and
>> short diphthongs were lengthened in late OE before such
>> voiced homorganic clusters as /ld/, so Anglian /kald/ >
>> late /ka:ld/ > ME /kO:ld/ except in the North.

> If PWGmc */a/ > early OE /æ/, then were did OE /a/ come
> from?

So-called restoration of [A]: stressed /æ/ was retracted to
/a/ (realized as [A]) before a back vowel in the following
syllable.

> Furthermore, what sort of conditioning factor could /lC/,
> /rC/, and /x/ (and only those three) have in common for
> affecting the vowel /æ/?

And the other front vowels as well. The answer is backness.
There's a nice discussion in Roger Lass, _Old English: A
Historical Linguistic Companion_, §3.6.2.

Brian