🚀 go-pugleaf

From article <1992Oct30.002351.11750@athena.cs.uga.edu>, by mcovingt@aisun3.ai.uga.edu (Michael Covington):
> In article <Oct.29.12.47.57.1992.19393@pilot.njin.net> hubey@pilot.njin.net (Hubey) writes:
>>
>>What are the necessary and sufficient conditions for a string of
>>letters to constitute a syllable ?
> 
> Surely you mean "string of _sounds_".  I'm not a phonologist, but here
> goes...
> 
> A syllable is a sonority peak.  That is, it is a local maximum of sonority.
> The middle of a syllable is surrounded by less sonorous sounds on both sides.
> 
> Sonority is defined by the Aissen/Hankamer sonority hierarchy, which I
> dimly remember as like this: 
>       Highest:  Vowels
>       Second:   Liquids (l, r)
>       Third:    Nasals (n, m)
>       Fourth:   Any other voiced continuant
>       Fifth:    Voiceless continuants
>       Bottom:   Stops
> 
> So when you say "bottle" you may well only be pronouncing /batl/ but the l
> on the end is a sonority peak (more sonorous than the t preceding or the
> silence following); it's a syllable without a vowel.
> 
> NOW THEN. If you are asking how to _divide_ a word into syllables, so that
> every sound belongs to a specific syllable, then you've opened an
> interesting can of worms.  How would you _know_ that, for instance,
> "refer" is "re-fer" rather than "ref-er"?  Hard to tell!
> 
> 
> -- 
> :- Michael A. Covington     internet mcovingt@uga.cc.uga.edu 
> :- Artificial Intelligence Programs       phone 706 542-0359
> :- The University of Georgia                fax 706 542-0349
> :- Athens, Georgia 30602 U.S.A.          amateur radio N4TMI

There is a nice theory that Clements put  out on sonority and the
syllable.  Instead of chopping up fricatives, stops continuants, etc...
using some different features (such as vocoid), a scale  such as 
O<N<L<G<W seems  to suffice.  (Obstruents, Nasals, Liquids, Glides,
and vowels.) 
    In any case, with the exceptions of peaks and reversals, one can
tell how MANY syllables there  are  in a word, by the fact that 
the syllable has a peak, which is +syllabic.  This can be syllablic
consonants, or a vowel, or whatever.  

	I.e.
		b l ai n d 


	V|        .   
	G|
	L|      .  
	N|            .
	O| .            .
	   b    l  ai n d

The peak here  is [ai].  This method done with multisyllabic words
show the  number of peaks, thus the number of syllables.   Like  he
mentioned above, there  is not always a clear distinction, as  language
dependent differences may change whats going on.

There is a  good  way (as a  general  principle), to figure syllables
though.  Onset creation rules.   The below one  is taken from  
Clements' 'The Role  of  the sonority cycle in  core  syllabification'

CSP:

a.  Associate each  [+syllabic ] segment to a syllable node
b.  Given P (an  unsyllabified segment) preceeding Q  (a  syllabified
    segment), adjoin  P to the syllable  containing Q  iff P has 
    a lower sonority rank than Q (apply iteratively)
c.  Given Q (a syllabified segment)  followed by R  (an unsyllabified
    segment), adjoin R to the syllable containing  Q iff R has a 
    lower sonority rank than Q  (apply iteraively)

where  the  order is important,  a, then b, then c.



take the word  "control"  (/k^ntrol/)    ^  =  schwa  or carrot .



a. 	o	o
	| 	| 
	^       o

b.	o	o
       /|     //|
     /  |   / / |
    k   ^  t r  o

k  is an  obstruent, lower in sonority rank than vowel ^
r is  a liquid,  lower in sonority rank than vowel o
t is  an obstruent, lower in sonority rank than liquid /r/ 

(Note,  /n/ doesn't get  connected going to the right, as 
it  is before /t/, so /n/ is greater in sonority rank than  /t/

c. 	o        o
      / | \    / | \ \
    /   |  \  /  |  \ \
   k    ^  n  t  r  o  l 

Now, here  you go rightward, connecting  the remaining  segments.

/n/ is  lower in sonority than  vowel  ^
/l/ is  lower in sonority than vowel  o

so you get, con-trol


It's not fool proof, but its pretty good.   There  are  all kinds
of  situations though.   Like  I said, plateaus, reversals, etc...
There are also different explanations  for them.   

Anyways, I've  been studying phonology for a while  now, and the
syllable and sonority is  one  the  most interesting things in my
opinion.


			Boy, did  I ramble on.

					Dave 


-- 
			     David Hahn
 		   University of Wisconsin - Milwaukee
		         hahn@csd4.csd.uwm.edu