Re: Unilang: the Phonology

On Wed, 18 Apr 2001 08:42:19 -0600, dirk elzinga
<dirk.elzinga@...> wrote:

>On Wed, 18 Apr 2001, David Peterson wrote:
>
>> Sorry.  I saw those +/- tables and just skipped right by
>> them.  I think that way of classifying sounds is quite
>> possibly the worst way every invented.  I was just
>> introduced to it this semester.  There are dozens of sounds
>> which can't be distinguished with that system.  I think the
>> IPA is much simpler. Or just saying things in plain
>> "English" (insert your natural language--or any other
>> language you prefer--in between the quote marks).
>
>Note that the alphabet of the IPA also contains implicit
>assumptions concerning the nature and organization of
>phonological elements--the most potentially damaging of them
>that there are in fact such things as segments. The segment is
>at best an entity derived from more basic elements--whether you
>want to represent them via distinctive features, atoms,
>particles, elements, or what have you. (There are some
>conventions used in the alphabet of the IPA which seem to
>recognize that segments are not primitive linguistic elements--
>such as the devoicing or nasalization marks--but in general the
>segmental bias of the scheme is hard to overcome.)
Thank you for making that point; that's one thing cleared. My own reason
for using SPE is because that allows me to define ranges without having to
recount all the possible values within the range. Using IPA for the scheme
I'm proposing there would be contrary to the whole idea; the idea is not to
assign absolutes, but merely assign boundaries, and describe what is
significant and what not. The SPE system is much more adequate for that.

Óskar