Re: C'ali update: tonal phonology

From:         Jörg Rhiemeier <joerg_rhiemeier@...>
> > Some of the following processes are apparent from even this
> > small data set:
> > (1) Leftward spread: prefixes acquire tone of root.
> > (2) Word-final L: all final syllables must be L.
>
> Yes, I have noticed that.
>
> > (3) No short contour tones.
>
> I see no long contour tones in your examples, either.
That's correct.  When I writing that post in the wee hours of the morning,
I didn't feel like exemplifying each rule, but rather just wanted to
mention it.  But I see that it may have been better to show how that
works.

> > Underlyingly, there is a contrast between
> > long and short /u/~/u:/ and long and short /o/~/o:/.
>
> One moment... I thought C'ali had no vowel length distinction, or
> did you change your mind (and your design)?
Also correct.  I changed my mind, basically because I thought the rule
was neat, but none of the words I already have distinguish long and
short vowels, so this was a handy way to do it.

> When a tone spreads
> onto an underlyingly long /u:/ or /o:/, it may spread onto the adjacent
> mora, but not the further mora. Now, the contrast of underlying short
> /u/ and /o/ with /y/ and /ö/ is always neutralized to [y] and [ö]
> respectively, while their long counterparts shorten.
>
> Hence,
>
> /u/ > /y/ /o/ > /ö/ /u:/ > /u/ /o:/ > /o/
>
> Correct?
Yes, there's a chain shift rule-relationship: you have to front the
short vowels first before you shorten the long vowels, otherwise there
wouldn't be any back rounded vowels at all!  I'm thinking about
whether I should restrict this chain shift only to certain phonological
environments (say, to closed syllables), or to make it morphophonological.
Thus, if this change is made, some underlyingly short /o/s might surface
as back vowels because they aren't triggered by the right environment.
It is also the case that the underlyingly long vowels, even after
shortening, are phonetically longer than any surface [y] are [ö],
even those not derived from back vowels.

There are, in other words, still details to be fleshed out.

> In this circumstance,
> only underlyingly long /u:/ and /o:/ receive surface contour tones;
> their fronted, underlyingly short counterparts may not do so.
>
> As I said above, I haven't seen any contour tones in your examples. Each
> syllable is either H or L. Where are the contour tones?
So, basically, if you have an underlying long vowel and it is followed
by a depressor consonant, then you could have a situation where the
second mora of the vowel is lowered, but not the first:

             HH L H     H                   H  H  HL H   L
   /t|e-t'ærro:-mpokw-qwo:/    =>         t|et'ærrompöqqwo
   thither-release-bird-2SgA
   'You are bird-releasing [as in a religious ceremony]'

Whether you see this as spreading of L and then deleting of that
L, or simply deleting the adjacent H, it is still the case that
these consonants create the lowered contour (phonetically just
marked by a downstep when the following syllable is H).

 =========================================================================
Thomas Wier	       "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637