On Saturday, October 19, 2002, at 08:26 , Peter Clark wrote:
> Quoting Yoon Ha Lee <[log in to unmask]>:
>> It's somewhat off the beaten path, but Korean has a number of mutations.
>> I can dig up my reference and post a few if you're interested.  :-)
>    Yes, please do so. The farthest I've gotten in Korean is teaching
> myself
> hangul. :)
Hey, you're way ahead of me in Welsh.  Er...

Anyway, here are bunches of phonological processes, from _The Korean
Language_ by Iksop Lee & S. Robert Ramsey.  (After reading this, the
differences and rationales for McCune Reischauer vs. Yale romanization
finally became clear to me.)  Er...Cymri?  They're probably not all
technically mutations (most of them seem phonologically predictable), but
maybe they'll give you some ideas.

I'm using [i"] for barred-i throughout.

nasal assimilation:
(1) kwuk-mul -> kwung-mul [k] -> [N]
(2) put-nunta -> pun-nunta [t] -> [n]
(3) eps-nun -> epnun -> emnun [ps] -> [m]
(4) khal-nal -> khallal [n] -> [l]

Weirdnesses with [l]
(5) sayngsan-lyang -> sayngsannyang [r] -> [n]
(6) tok.lip -> tongnip [k] -> [N], [l] -> [n]

(7) path i -> pach i [t_h] -> [tS_h]
(8) mitat.i -> mitaci [t] -> [c_h]
(9) tat.hita -> tachita [th] -> [tS_h]

Labial assimilation:
The unrounded vowel [i"] barred-i becomes [u] after a labial consonant [p]
, [p_h], [p']+, [m], or after the semivowel [w].

+ Korean "tensified" stops.  If another language exists that makes use of
them I haven't been able to find it...though if someone knows of one, by
all means enlighten me.  :-)

disappearing h: between vowels, or between vowel and [j], [w], [m], [n],
   or [l], [h] will disappear
disappearing w: after [p], [p_h], [p'], [m], [o], or [u], [w] drops
reduction of cye, chye, ccye: [cj@], [c_hj@], [c'j@] are pronounced all the
dropping of l before t, n, s or c: in some compounds, [l] drops before [t]
   [n], [s] or [c]
reinforcement: when 2 lax (non-tensified) consonants occur together within
   a word, the 2nd is tensified

(Skipping "genitive particle s" because the explanation is long...)

Verb stem weirdnesses:
(1) noh.ko -> no.kho [hk] -> [k_h]
(2) noh.teni -> no.theni [ht] -> [t_h]
(3) -> no.chi [hc] => [tS_h]
(4) -> no.sso [hs] -> [s'] (tensified [s])
(5) noh.nunta -> non.nunta [hn] -> [nn]

final -p/-w: One type of -p at the end of a verb stem weakens to -w before
   a vowel (the other type doesn't change at all):
-p/-w: nwup/nwuw- [nup]/[nuw] "lie down"
        nwup.supnita [nupsi"mnida]
        nwup.ko [nupk'o]
        nwuwe [nuw@]

final -t/-l: one type of -t at the end of a verb stem weakens to a -l [r]
   before a vowel (the other kind doesn't change at all):
-t/-l: tut- "listen" [ti"t]
        tut.supnita [ti"tsi"mnida]
        tut.ko [ti"tk'o]
        tul.e [ti"r@]

dropping of final -s: -s at the end of some stems drops before a vowel (the
   other kind doesn't change at all):
-(s): pu(s)- "swell"
       pus.supnita [bi":s'i"mnida]
       put.ko [bi":k'o]
       pu.e [bi"@]

l-stems: generally convoluted and irregular:
(1) drops before [n], [l], [p], [s] but appears before other consonants
     or a vowel
     e.g. tul.ta [ti"lta] "cost":
 [ti"na], tup.nita [ti"mnida], tu.sita [ti"Sida],
          tu.sey [ti"se], tul.ko [ti"lko], tul.e [ti"r@]...
(2) -lu- verb stems where u [i"] drops and l [r] doubles to [ll] with the
     infinitive ending -e [@] or the past marker -ess [@t']/[@s']
     e.g. hulu.ta [hi"ri"da] "flow"
 [hi"ri"na], hulup.nita [hi"ri"mnida], hulu.ko [hi"ri"go],
 [hi"ri"dZi], hulle [hi"ll@], hullessta [hi"ll@t'a]

(3) Other verb stems ending in -lu-: when followed by the infinitive
ending -e [@] or the past marker -ess- [@s']/[@t'] the u [i"] doesn't drop,
  and another l [r] is added
     e.g. phulu.ta [p_hi"ri"da] "be blue"
 [p_hi"ri"na], phulup.nita [p_hi"ri"mnida],
          phulu.ko [p_hi"ri"go], [p_hi"ri"dZi], phulu.le [p_hi"ri"
          phulu.lessta [p_hi"ri"r@t'a]

Okay.  That was probably more than *anyone* ever wanted to know about
Korean phonological processes....

>> I envy you, actually.  I have a Welsh grammar somewhere but darned if I
>> can find it.
>     Next time you're in Minnesota I'll lend you my "Teach Yourself Welsh,
> "
> although you should not that it's not the best--I'm especially peeved that
> there's no IPA description of the sounds. Grr...

Yeah, I hate that about "teach yourself" books sometimes.  *sigh*  Now all
I have to do is figure out how to get to  Minnesota.  ;-)  (I'm in Boston

Yoon Ha Lee [[log in to unmask]]

Even good companies make mistakes.  Great companies fix them.--Steve Jobs