[Spellyans] reDistribution of <i> and <y>
everson at evertype.com
Thu Jul 24 23:47:50 IST 2008
At 23:48 +0300 2008-07-24, Owen Cook wrote:
>He proposed two things. I've already said I
>agree with you that his proposal vis-à-vis
>unstressed syllables would not be workable or
>desirable. But in stressed syllables, he
>proposes (if I understand him right) to allow
><i> to be either short or long. This means any
><y> in closed syllables will itself signal an
>alternation with RLC <e>. Hence splitting by 50%
>the need and usefulness of the diaeresis.
It DOESN'T WORK.
It just shifts the problem. Now we have
<Crist>|<kyst>|<bys>~<bes>|<res>. We mark <bÿs>~<bës>
If you make the change you propose, you simply shift the problem:
<Crist>|<kìst>|<bys>~<bes>|<res>. We would STILL
have to mark <bÿs>~<bës>, or at least <bës>,
because of the <bes>|<res> clash, and moreover
you've added <ì> where we didn't need it before
AND you've created a big delta from the SWF (and
UC, UCR, and KK in terms of words like <kyst>.
> > On top of that he wants to use <y> now as an
>> umbrella graph for RMC /i/ and RLC /e/
>Stop it, Michael. This is utter nonsense. Look
>at his email. There is absolutely no proposal of
>getting rid of RLC <e>. His proposal is exactly
>NOT to use <y> (long or short) as an umbrella
>graph, but to use it in RMC only where RLC still
>has <e> (long or short). Repeat: still has <e>.
He said: "But where it's short, I would like to
write <i> where RLC also has /i/ rather than /e/
and only spell them <y> where RMC has /i/ and RLC
I read that as using an umbrella graph for short
RLC /e/. I opposed <y> as an umbrella graph for
long [e:] and I also do for short [E] or the
> > I'll say it again: we fought for and won the
>> <bys>~<bes> distiction.
>And I'll say it again: Dan's proposal does not
>touch the <bys>~<bes> distinction. Indeed, it
>extends it to those words in which there is /I ~
>E/ alternation (I've found about half a dozen).
I would extend it to <pÿsk>~<pësk> which have
short vowels but there are very few of those
> > Right now we are chasing our tails and failing to
>> come to consensus.
>Look, we're not going to come to consensus if
>alternate proposals aren't given an airing.
At what point does an issue become closed? So far
I'm NOT getting anywhere writing the grammar.
Until we've got that we've got no orthography to
> > Owen, we cannot throw out the
>> whole thing and go to spell gwinn/gwidn when the
>> SWF offers us gwynn/gwydn for this. This is a
>> complete overturning of EVERYTHING. You're
>> putting me back to Square One,
>I really think you're overreacting. A complete overturning of everything?
UC has <gwyn gwyn>. SWF <gwin gwynn> is an improvement.
> Come on -- how often does MC /I/ remain /I/ in
>Late Cornish? I can find about a dozen such
>monosyllabic words of Cornish origin, and about
>an equal number of English borrowings normally
>spelt with <i> anyway in modern English. Truth
>is, there just aren't that many words like
The distribution of <i> for long [i:] and <y> for
short [I] in stressed monosyllables is at the
heart of the SWF. There is no problem with it.
There is nothing problematic about writing <jyn>
or <gwynn>~<gwydn>, AND those forms are familiar
to everybody using UC and KK. Why change these?
If I understand what's proposed:
I think KS1 here is more coherent, closer to the
SWF, and more likely to succeed.
> > If I understand what Owen has
>> written, he'll propose to change the first to
>> <kìst>. That just shifts the "diacritic problem"
>> to another place.
>Certainly, but as you've admitted yourself,
>Michael, there are a lot of words that have /e:
>~ i:/ alternation (I count 19 items with <ei> in
>the KS 16 document). If we follow your proposal,
>all of these words need diaereses.
Nicholas (who is doing a fair bit of translation
right now -- I won't tell you what!) says that in
practice it's not so many words. There are a
number of common words, but the others are not so
common. The look of a text isn't drastically
>How many words have anomalous short /I/ that
>remains stable in Late Cornish? Maybe half a
That's saying a lot. But I don't see where your statement leads.
> > The remaining deltas have to do with <i>
>> and <y> in polysyllabic loanwords, and then with
>> a certain number of other "etymological"
>> spellings "inherited" from KK. These should be
>> reviewed in a systematic fashion.
>Yes, and I guarantee you, if we fix them in a
>systematic fashion, these will affect a good
>deal more than my two-dozen words.
Maybe, but NOT the orthography. Those are
morphological questions. If we end up (because of
research, not just reconstruction) really having
a regular declension class that has
<taves>/<tavosow> as one of its members, on solid
linguistic grounds, that is one thing. That's not
the same thing as whether 'white' is spelt
<gwynn> or <gwinn>.
>Consider it. This isn't a totally bad idea. (And
>it does serve one thing on KS's to-do list --
>mildly, and systematically, increasing the
>occurrence of <i> with respect to <y> to satisfy
>RLC users' aesthetic sense.)
I think that our using <i> regularly in initial
position will go a long way to helping that. But
the system has to be coherent. Marking the
bÿs/bës distinction is one thing; it solves the
clash with bys/res. The proposal doesn't solve
that; it just leads to bìs/bës or bìs/bÿs/bës.
Michael Everson * http://www.evertype.com
More information about the Spellyans