--- In qalam@yahoogroups.com, "Richard Wordingham"
<richard.wordingham@...> wrote:

>
> --- In qalam@yahoogroups.com, "suzmccarth" <suzmccarth@...> wrote:
> > The problem here is that the precomposed characters, the

syllabics

> > plus diacritic, do not represent a widespread indigenous use.
> There
> > was once a Cree elder known for using diacritics phonemically.
> This
> > skill was recognized as a skill particular to his family, his
> > specialty. It was remarked on as unusual. One family!
>
> Isn't this akin to the Cyrillic yo-letter U+0401/U+0451 (like

Roman

> ë) in Russian? This letter is used chiefly for foreign

learners.

> There is also the less significant use of accents in English, both
> foreign (e.g. fiancée) and native, especially poetical,
> e.g. 'hornèd' and 'learnèd'.

It was the foreign linguists who decided that diacritics should be
used phonemically (obligatory and consistent) because it is easier
for them.

>I could be more brutal and say that it
> sounds from your account as though most Cree can't spell.

What a bizarse comment. First, the Cree had *never* established the
use of obligatory diacritics for themselves. Indigenous Cree
dotting is usually compared to Hebrew partially marked text. If you
can't produce fully marked Hebrew text does that mean you can't
spell? I guess I don't know the answer to that. I would be
interested to hear what the expectations are for Hebrew. But it was
never a standard for Cree. Partially marked was used but at the
writer's discretion.

Are Hebrew dictionaries collated by letter *and* vowel marking? I
would be interested to hear about this too. If Hebrew letters were
encoded separately for each possible vowel mark and combination of
markings wouldn't that create multiple encodings for each word as
people could produce unmarked, partially marked and fully marked
spellings? I can't even find any marked Hebrew text on the internet.

>I could
> compare it to my name being misspelt <r><i><ch><a:><silent d><r>

in

> Thai, where the 'diacritic' pattern of <i>..<silent> is right, but
> unmarkedly silent <r> has been put in its commonest place. It may
> be common, but it's still WRONG.

There is no WRONG and RIGHT about choosing to dot or not in Cree.
Dottings are added to the full syllabics at the writer's
discretion. Why create quadrupled codecharts? Actually I know who
created the system in the eighties (for a bilingual dictionary) but
I don't know how this encoding fits into Unicode principles or what
pragmatic effects result from it.

I am not saying that Unicode charts force a certain collation but I
am saying that these different collation sequences now exist.

> As to the charts, if someone were transcribing my English
> handwriting, I hope they wouldn't look at the Unicode Latin charts
> and carefully work out which letters I had dotted and record the
> result this. If I write <undotted i><n overdot> when writing
> English, I mean <i><n>. Recording it as <undotted i><n overdot>
> would be gratuitously confusing.

Irrelevant.

>
> I may have misunderstood the brief accounts I've seen, but doesn't
> Nynorsk result in multiple spelling for a great many words?
>
> Richard.

Suzanne