Re: vowel harmony

      --- In conlang@yahoogroups.com, Tristan Mc Leay <conlang@T...>
  > wrote:
> On Mon, 2005-11-21 at 19:35 +0000, tomhchappell wrote:
>> [snip]
  >> .... Front vs. Back frequently has at least three, and sometimes
  >> has more than three values;
>
> Really? I thought having even three values was pretty rare, well,
  > except phonetically. But I'm of the impression that phonemically the
  > third value is almost always not relevant and so it'd be invisible to
  > vowel harmony. So a language with [i] [u\] [u] would have /i y/ as
  > front and /u/ as back; or one with [i] [i\] [u] would have /i/ as front
  > and  /i\ u/ as back.
> And certainly I was of the understanding that three was an
  > absolute upper limit, and *no* language had more than three
  > backness values.
> (Excepting when it's actually something like a tense-lax
  > distinction that is concomitant with a height distinction.)

  --- In conlang@yahoogroups.com, caotope <johnvertical@H...>
  > wrote:
> tomhchappell wrote:
  >> [snip]
>> Front vs. Back frequently has at least three, and sometimes has
  >> more than three values;
>
> It's been asked already, but you are here talking phonetically, not
  > phonemically, right? /a/ is quite common of course, but I've thought
  > it usually analyzes as front (whenever that is relevant anyway)

  I am afraid that I do not understand enough about the details of your
  questions, where they concern the differences between phonemics
  and phonetics and phonology, to see how the following examples
  answer them.

  However, the following examples are among the reasons I thought
  frontness/backness of vowel phonemes might have more than three
  values.

  They all come from The Cambridge Encyclopedia of Language,
  Second Edition, by David Crystal.
  Cambridge University Press 1997
  ISBN 0-521-55967-7
  P29.C64 1997
  Part IV:  The Medium of Language -- Speaking and Listening
  Section 28:  The Linguistic Use of Sound
  Topic:  Comparative Phonology
  Page 169

  -----
  Panjabi, an Indo-European Language, has a 20-vowel system.
  They occupy 5 levels of Frontness/Backness,
  and 6 levels of Closeness/Openness.
  It has these Front vowels:  /i/  /i~/  /e/  /e~/  /&/  /&~/  /a/  /a~/
  It has these Near-Front vowels:  /I/  /I~/
  It has these Central vowels:  /@/  /@~/
  It has these Near-Back vowels:  /U/  /U~/
  It has these Back vowels:  /u/  /u~/  /o/  /o~/  /Q/  /Q~/
  -----
  English, an Indo-European Language, has a 12-vowel system.
  They occupy 5 levels of Frontness/Backness,
  and 7 levels of Closeness/Openness.
  It has these Front vowels:  /i/  /e/  /&/
  It has this Near-Front vowel:  /I/
  It has these Central vowels:  /@/  /3/
  It has this Near-Back vowel:  /U/
  It has these Back vowels:  /u/  /V/  /O/  /A/  /Q/
  -----
  Masai, a Nilo-Saharan language, has a 9-vowel system.
  They occupy 4 levels of Frontness/Backness,
  and 5 levels of Closeness/Openness.
  It has these Front vowels:  /i/  /e/  /E/  /a/
  It has this Near-Front vowel:  /I/
  It has this Near-Back vowel:  /U/
  It has these Back vowels:  /u/  /o/  /O/
  -----
  Kunama, a Nilo-Saharan language, has a 7-vowel system.
  They occupy 4 levels of Frontness/Backness,
  and 4 levels of Closeness/Openness
  (thus are a counterexample to a "universal" that every vowel system
  has more levels of Closeness/Openness than of Frontness/Backness)
  It has these Front vowels:  /i/  /e/  /a/
  It has this Near-Front vowel:  /I/
  It has this Near-Back vowel:  /U/
  It has these Back vowels:  /u/  /o/

  ----------

  So, anyway, those were some of my reasons for thinking that
  languages could have sets of vowel phonemes having more than three
  distinct levels of frontness and backness.  Granted, none of those
  examples has a set of four vowels all having the same value of
  closeness/openness, roundedness, nasality, and tongue-root
  position, distinguished only by frontness/backness; indeed, among
  those languages I have just cited, no more than two vowels have all
  other features equal except frontness/backness. But I don't know if
  that proves Tristan's point, or answers John's question; I really didn't
  understand them.

  I'd be happy to have someone explain it to me --
  offlist or on, as you prefer.

  Thanks,

  Tom H.C. in MI





		
---------------------------------
 Yahoo! Music Unlimited - Access over 1 million songs. Try it free.

Reply