Re: another odd form

Hi Ole,

I have number of beginner's question if you don't mind. I hope it's
at least clear what I'm asking, even if the questions might make you
want to exclaim (as in H Smith's marginalia) åsne!

Going back to your original post:

>Originally the /u/ of /upa/ was dropped for
>prosodical reasons,

if /upa/ ends up going to /para/ what is gained prosodically? Same
syllable count and quantity, right?

Do you mean that there was an intermediate form /patthaddho/ (with
one less total syllables, fulfilling som now unattested prosodic
requirement) that existed for a while and then was expanded with the
glides?

>and speakers inserted compensatory glides.

What is a compensatory glide and why is there a perceived need for
it? To restore the original syllable count (or rhythm)?  What does it
compensate?

>  /r/ is a
>highly frequent glide in Ja, occuring in intervocalic position. According to
>my calculations is represnts one third of all glides in Ja V /m/
>representing two thirds of the occurrencies. Since Pali disallows the
>cluster /rtth/ a vocalic on-glide /a/.

Aren't glides normally used to prevent vowel hiatus (or as you said,
in intervocalic positions)? If so, if the order of events is
insertion of /r/ glide, and THEN the /a/ on-glide to avoid rtth,
doesn't that leave no reason for the r glide to arise in the first
place?

Or do you mean that the /r/ glide is just a result of the speech
organs transitioning from the /pa-/ towards the /-ttha/ so it could
arise and then force the on-glide /a/ ?

As you can see, I don't quite follow the chain of events here. Any
help would be appreciated. I'm interested in this sort of reasoning,
but know very little about it. It's probably time to read the
phonetics section in geiger which I skipped (apart from the law of
morae that I had to know for exams)

best regards,

/Rett