H. S. Teoh wrote:

>Me too. In fact, I even have a problem understanding what some IPA sounds
>are... The laterals and trills completely elude me thus far (except for
>the simplest laterals like the English l).

You are not alone. I've done all the compulsory phonetics of a linguist
training to do field work on an undescribed language, and I still get lost
over laterals and trills. I'm doing a Field Methods course (trains us how
to do the actual field work) on Pima. I have found myself transcribing a
tap (like the r in Spanish), an English l, an English r, and a retroflex l
(combine English l and r and you've got the idea). 10 hours of work with
our consultant and fifteen collaborating students later, we *think* that
these are all the same sound. I was conservative, BTW -- one students
transcribed this sound seven different ways. And some of these students
intend to become professional phoneticists.

>  Then there are the unvoiced
>nasals, which I just can't figure out.

You just have to hear them.

>  In general, though, consonants
>cause me less grief than vowels... I just do not *get* the IPA vowel
>system. It seems to make fine distinctions in what are allophones to me,
>and not-fine-enough distinctions in what I consider heterophones.

You have to understand what IPA is designed for. It is not capable of
giving very precise transcriptions of a sound. The symbols are only an
approximation that can be used to show contrastive sounds. The more common
two sounds are contrasted in the world's languages, the more likely they
have a separate symbol. Less common sounds get marked by the complex
diacritic system.

I am a bit perplexed about your "allophone" comment though. IPA is supposed
to be language independant. Whether or not something is an allophone in a
particular language is completely irrelevant to the system: they are not
allophones in all languages.

Marcus Smith
AIM:  Anaakoot
"When you lose a language, it's like
dropping a bomb on a museum."
   -- Kenneth Hale