Re: Question about a grammatical term
From: | Roger Mills <romilly@...> |
Date: | Saturday, October 5, 2002, 0:02 |
Padraic Brown wrote:
>
>> "Secondary stress" is a conventional concept.
>
>How many levels are "conventional"?
>
As I recall, at least 3 in the Chomsky-Halle schema, and maybe even 4-- the
latter for very reduced (surface) phenomena like [t@] 'to', [@n] 'and',
though I'm not sure about that.
1 of course is main stress: yellow l3, jacket l3
3 is ordinary unstressed; vowels may reduce to [@]
2 is a demoted 1, in compounds, phrases, and long words: yellow-jacket
(wasp) 13 23 and yellow jacket (adj N) 23 13
Seems to me there was something about an alternating stress pattern too, to
account for the unreduced vowels in such things as:
telegraph 132 telegraphy [t@lEgr@fi] 313(2? 3?) telegraphic 2312
(Always struck me as odd that all the examples of this involved
tele-..........No, that's unfair-- it works for things like Mississippi
[mIs@sIpi] 231(2,3?) or stabilize 132. It does seem that it's the
non-native portion of the vocab. that throws everything out of whack.
Disclaimer: I am not a True Believer in this particular gospel.