Re: <ll> and <y> in Argentian Spanisg (was: Chinese languages or language?)

Ruud Harmsen wrote:
>
> Sat, 14 Jan 2006 15:11:08 GMT: "Peter T. Daniels"
> <grammatim@xxxxxxxxxxxxxxxx>: in sci.lang:
>
> >> >The IPA for the jiim-sound is [dZ];
> >>
> >> Not U+025f, voiced palatal stop ("barred dotless j")?
> >> (J in Krishenbaum, no symbol in Sampa).
> >
> >No, it most certainly is not a palatal stop. Do you not have any Modern
> >Standard Arabic songs in you collection of soundclips, that you could
> >consult since you don't believe things that you read?
>
> No songs. In this case I do rely on a written source. My earlier
> statement that I don't read books was a jocular exaggeration. I think
> I explained that later, and if not, I do now.
> My source is the 1943/1958 Teach Yourself (bought in a second-hand
> shop much later, possibly 1972). It lists shiin, raa', nuun and saad
> in a column headed "Gums", but jiim and yaa' under "Palatal".
>
> >> >one of the arguments against IPA is
> >> >its (non-)handling of affricates.
> >>
> >> http://www.ling.hf.ntnu.no/ipa/full/ipachart_other.html
> >> "Affricates and double articulations can be represented by two symbols
> >> joined by a tie bar if necessary."
> >
> >Which does not honestly represent a phonemic system;
>
> Of course not, how could it? A phonemic system only makes sense for a
> specific language under consideration. IPA provides a set of symbols
> for general use, for any language.

And the kludge it provides for affricates does not make sense for the
two specific languages under consideration.

> >the IPA's stated
> >goal is to provide _a symbol_ for every sound that's distinctive in some
> >language. (It isn't designed for extremely narrow phonetic
> >transcription, for instance.) Writing affricates with _"two symbols"_
> >does violence to that goal, and misrepresents (e.g.) the English and
> >Arabic phonemic systems.
>
> Simple solution: consider those the combination of two symbols as one
> symbol, adding the tie bar if necessary to avoid any unclarity.

No, "considering" two symbols to be one symbol is ridiculous. There's no
"if necessary" about it.

> >English and Arabic /j/ pattern like the other
> >stops, not like stop+fricative sequences.
>
> OK. That's one of the purpose of the tie bar, viz. to make that fact
> unambiguous for these languages.
> Another solution may be to write t and S or d and Z close together so
> they become a single ligature phonemic symbol.
> What's the problem?

That is how IPA _used to_ do it. It was very slightly less problematic,
because it was still obviously a sequence of two letters, rather than
one letter = one phoneme, as in their stated goal.
--
Peter T. Daniels grammatim@xxxxxxx
.

Relevant Pages